AI-103 Practice Question 47: Read OCR vs Tagging
Microsoft Azure AI Apps and Agents Developer Associate (AI-103) · Test 5 · question 7 of 10
A retail app lets shoppers photograph a shelf price tag. The app must return the printed price and product code as text so it can compare them against the online catalog. The team already tried Image Analysis tagging and object detection, which only return labels like "sign" and "text", not the price or code themselves.
What should the app call instead?
Show answer and explanation
Answer:
The Image Analysis 4.0 Read feature "offers the ability to extract printed or handwritten text from images in a unified performance-enhanced synchronous API." Requesting Read in the features parameter returns the actual recognized lines and words, including a price and a product code, as text with bounding boxes and confidence scores, which is exactly what comparing against a catalog requires.
- B is wrong because Smart Crop reframes or thumbnails an image around its most relevant region; it never reads the characters printed on the tag. - C is wrong because tags are a fixed vocabulary of scene and object labels. Loosening the confidence threshold returns more labels like "sign" or "retail," never the specific digits and characters printed on the tag. - D is wrong because dense captions generate natural-language descriptions of regions in the image ("a price tag on a shelf"). A caption can mention that a tag is present, but it doesn't reliably reproduce the exact price or code as characters the way OCR does.