Course (Part 606)

Course: Imported PDF Course

Composition "(no extractable text) (no extractable text)" is a notation commonly seen in digital documents to indicate that text within an image or scanned file cannot be selected, searched, or copied. This label highlights an accessibility and usability gap: users cannot extract textual content for quoting, indexing, translation, or assistive technologies. The root cause is often a document that contains raster images of pages rather than embedded character data. Optical character recognition (OCR) can sometimes convert such images into machine-readable text, but OCR accuracy varies with image quality, fonts, and language. Metadata or alternative text may be absent, leaving automated tools and screen readers without information to convey. For legal, archival, or collaborative workflows, "(no extractable text)" signals that extra processing steps are required before the document can be fully integrated. Publishers, librarians, and knowledge workers therefore prioritize creating born-digital files or performing careful OCR and proofreading. When sensitive content is present, simply exposing text can raise privacy and copyright concerns, requiring policy-aware handling. In many educational contexts, inaccessible texts hinder learners with disabilities and stretch instructor workload for redesigning materials. Automated translation systems also fail silently when they encounter non-extractable content, reducing cross-lingual access. A best practice combines preventive measures—producing accessible originals—and remedial measures—annotating or transcribing scanned content. Modern tools increasingly offer integrated OCR and tagging features, but human verification remains essential for reliability. Recognizing the label "(no extractable text)" should trigger a clear remediation workflow that balances technical, legal, and pedagogical considerations. Semantic HTML Lecture Sheet

Lesson

Topic: (no extractable text) (no extractable text)

Learning Outcomes

• Define what "(no extractable text)" indicates in a digital document.
• Identify common causes that produce non-extractable text.
• List remediation techniques such as OCR, transcription, and metadata tagging.
• Assess accessibility and legal implications of non-extractable content.

Conceptual Explanation

The phrase "(no extractable text)" marks that textual characters cannot be programmatically selected or searched because the content exists as an image or in a format without embedded character encoding. Causes include scanned pages, screenshots, or PDFs exported without text layers. Consequences range from reduced accessibility for screen-reader users to impeded indexing by search engines and inability to quote passages for research. Remediation strategies include performing OCR with verification, adding alternative text and metadata, or producing a manual transcript where accuracy is critical. Workflows must consider copyright, privacy, and institutional policies before converting or publishing extracted text. Tools that combine OCR, layout analysis, and language models can accelerate conversion, but human review ensures semantic fidelity, especially for specialized terminology or degraded images. Ultimately, creating born-digital, tagged documents is the most robust prevention against recurring occurrences of non-extractable text.

Bilingual (EN → বাংলা)

English	বাংলা (Bangla)
"(no extractable text)" indicates text cannot be copied or searched.	"(no extractable text)" নির্দেশ করে যে লেখা অনুলিপি বা অনুসন্ধান করা যাবে না।
Common causes: scanned images, screenshots, exported PDFs without layers.	সাধারণ কারণ: স্ক্যান করা ইমেজ, স্ক্রীনশট, টেক্সট লেয়ার ছাড়া PDF এক্সপোর্ট।
Remedies: OCR, transcription, adding metadata and alt text.	সমাধান: OCR, ট্রান্সক্রিপশন, মেটাডেটা এবং বিকল্প লেখা যোগ করা।

Important Notes

• Always verify OCR output; automated conversion often introduces errors.
• Respect copyright and privacy when extracting and publishing text.
• Provide transcripts or accessible alternatives for learners with disabilities.
• Prefer producing born-digital, tagged documents to avoid recurring issues.

Exercises → True / False / Fill in the Blanks / MCQ / Matching / Descriptive / Creative

True / False

1. Statement: A PDF with an embedded text layer typically does not display "(no extractable text)".
   Answer

   True. If a PDF contains an embedded selectable text layer, text is extractable and searchable.

2. Statement: OCR always produces perfect results regardless of image quality.
   Answer

   False. OCR accuracy depends on image quality, fonts, language, and layout complexity.

3. Statement: Adding alt text to an image automatically makes the entire document searchable.
   Answer

   False. Alt text helps accessibility but does not replace textual content for full-text search unless added as searchable metadata.

4. Statement: Non-extractable text can block automated translation services.
   Answer

   True. If the source text is not machine-readable, automated translation systems cannot access it directly.

Fill in the Blanks

1. Blank: The process of converting images of text into machine-readable text is called ________.
   Answer

   Answer: Optical Character Recognition (OCR).

2. Blank: Providing a manual ________ is essential when OCR cannot reliably recover content.
   Answer

   Answer: transcription.

3. Blank: Born-digital files are preferable because they include an embedded ________ layer for search and selection.
   Answer

   Answer: text (or text layer).

Multiple Choice Questions (MCQ)

1. Question: Which action most directly remedies "(no extractable text)" in a scanned document?
   1. Adding a watermark
   2. Running OCR and proofreading results
   3. Compressing the image
   4. Converting to a different image format
   Answer

   Answer: B. Running OCR and proofreading results.

2. Question: Which of the following is NOT a consequence of non-extractable text?
   1. Reduced screen-reader accessibility
   2. Improved machine translation quality
   3. Difficulty quoting passages for research
   4. Problems with search engine indexing
   Answer

   Answer: B. Improved machine translation quality (this is not a consequence).

3. Question: Which metadata element can help make scanned documents more usable without full OCR?
   1. Title and descriptive tags
   2. Pixel resolution only
   3. Compression ratio
   4. Embedded fonts
   Answer

   Answer: A. Title and descriptive tags.

Matching

Match the left column (issue) with the right column (remedy).

Issue	Remedy
A. Screen-reader cannot access content	1. Add alt text and provide transcripts
B. Text in image is low-quality	2. Enhance image and run OCR, then proofread
C. Document missing searchable metadata	3. Add descriptive metadata and tags

Answer

Answer: A → 1, B → 2, C → 3.

Descriptive

1. Prompt: Describe a 3-step institutional workflow to process archival PDFs labeled "(no extractable text)".
   Answer

   Sample answer: Step 1 — Ingest and catalog the files with metadata indicating format and condition. Step 2 — Run a staged OCR process prioritizing high-quality images, followed by human verification and correction of critical passages. Step 3 — Attach transcripts, searchable text layers, and access controls; update the catalog to reflect new searchable status while ensuring copyright and privacy compliance.

Creative

1. Prompt: Propose an interface feature that helps users immediately identify and remediate non-extractable text in a document repository.
   Answer

   Sample idea: Implement a "Text Accessibility" badge on each file card that shows status (Extractable / Image-only / Needs Review). Clicking the badge opens a remediation panel offering one-click OCR with configurable language settings, a link to request human transcription, and a preview of detected text with inline edit and save-to-repository functionality. The panel would also surface copyright checklists and a confidence score to guide prioritization.

Activities, Vocabulary, Homework, Practice Task

Activities

1. Activity: Collect three PDFs labeled "(no extractable text)" and run OCR; compare results and note common errors.
   Answer

   Expected output: A short report summarizing OCR accuracy, error types (misrecognized characters, line breaks), and suggested corrections.

2. Activity: Create an accessible resource page that explains why non-extractable text matters to different stakeholders.
   Answer

   Expected output: A one-page guide with sections for researchers, librarians, and learners with disabilities.

Vocabulary

• OCR — Optical Character Recognition.
  Answer

  Definition: A technology that converts images of text into machine-readable text.

• Born-digital.
  Answer

  Definition: Content originally created in a digital form with selectable text and metadata.

• Alt text.
  Answer

  Definition: Short textual descriptions of images to aid accessibility.

Homework

1. Homework: Prepare a 300–500 word plan describing how your department would prioritize converting legacy scanned documents to searchable formats.
   Answer

   Guidance: Include criteria for prioritization (usage frequency, legal value, accessibility needs), estimated resources, and a quality assurance step for OCR outputs.

Practice Task

1. Task: Take one short scanned page, run OCR, correct the output, and submit original and corrected text alongside a brief reflection on difficulties encountered.
   Answer

   Expected submission: Original image, OCR text with annotations for corrections, final corrected text, and a 150–200 word reflection.

End of lesson: Review important notes and ensure remediation steps respect legal and accessibility requirements.