6.1.2. Extracting Information from Images
💡 First Principle: Image extraction applies the same analyzer idea to photos and scans: read text in the image (OCR is part of it) and pull out structured fields or content. A photo of a business card yields name, title, and phone; a shelf photo can yield product details. The analyzer defines what to look for.
This overlaps with computer vision but is aimed at structured output: the goal isn't just "classify this image" but "give me these specific pieces of information from it." Content Understanding handles the OCR plus the field identification in one step, returning data your application can act on directly.
⚠️ Exam Trap: Image extraction and image classification differ in output. Classification gives a label ("this is a receipt"); extraction gives the values ("merchant: Café X, total: 12.40"). When a scenario needs the data inside the image, that's extraction.
Reflection Question: A pharmacy wants to read prescription details from a photo of a label. Is classification, OCR alone, or information extraction the right fit, and why?
