AI-900Domain 3: Describe Features of Computer Vision Workloads on AzureComputer VisioneasySingle choice
A mobile app scans whiteboards and extracts the handwritten text into editable documents. Which capability is used?
A.Image captioning
Captioning describes scenes in words — doesn't extract written text.
B.Optical character recognition (OCR / Read API)
Text extraction from images = OCR. Correct.
C.Facial recognition
Facial recognition is identity — unrelated.
D.Semantic segmentation
Segmentation labels pixels — not text extraction.
Why this answer
OCR converts images of text (printed or handwritten) into machine-readable text — Azure AI Vision's Read API is optimized for documents/handwriting vs the basic OCR for small embedded text.
Correct answer: B · Last reviewed 2026-02-14
Official AWS references
Practice this interactively
Answer without seeing the solution first in the Computer Vision practice session.
Practice Computer Vision