AI-900 Practice Question: Describe features of computer vision workloads on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✓

  Optical Character Recognition (OCR)

  Why this is correct

  Optical Character Recognition (OCR) is the correct choice because the Azure AI Computer Vision OCR/Read API is specifically built to extract printed and handwritten text from images and documents, returning machine-readable character strings. Unlike general image analysis, OCR focuses entirely on localizing and recognizing text, producing line-level bounding boxes, confidence scores, and raw text output. For a logistics company digitizing thousands of handwritten shipping labels, OCR is the prebuilt, ready-to-use service that directly satisfies the text-extraction requirement without custom training.

• ✗

  Image Analysis

  Why it's wrong here

  Image Analysis is incorrect for this task because it is designed to describe the overall content of an image—such as visible objects, scene types, and tags—rather than to transcribe text characters. While Image Analysis can detect that a label contains text as a visual attribute, it never outputs the actual string of handwritten characters, leaving the shipping information inaccessible. In contrast, the OCR capability in Azure Computer Vision is the dedicated tool for converting text pixels into usable data.

  When this WOULD be correct

  If the question asked for identifying objects or describing the content of an image (e.g., 'Which capability can detect that a photo contains a dog and a ball?'), Image Analysis would be correct.

• ✗

  Face detection

  Why it's wrong here

  Face detection is not applicable because its sole objective is to locate human faces within an image, returning bounding boxes and facial attributes like age or emotion, with no capability to read or interpret characters. In the shipping-label scenario, the target content is handwritten text, not faces, so face detection would completely ignore the information that needs to be digitized. Even if a label included a photo of a sender, the face-detection service would still skip the text, making it fundamentally unsuitable for the text-extraction workflow.

  When this WOULD be correct

  A question asking for a solution to detect and locate faces in images for security or demographic analysis, such as counting people in a crowd or verifying identity in a photo.

• ✗

  Custom Vision

  Why it's wrong here

  Incorrect because Custom Vision is used to train custom image classification or object detection models. While it could theoretically detect text regions, it is not optimized for text extraction and requires extensive training data, whereas OCR is a prebuilt, ready-to-use feature.

  When this WOULD be correct

  A company needs to identify specific objects (e.g., logos, package types) in shipping forms, and the available pre-built models do not cover those objects. Custom Vision would allow training a custom object detection model to identify those specific items.

Option-by-option analysis

Why each answer is right or wrong

Understanding why wrong answers are wrong — and when they would be correct — is what separates a 750 score from a 900. The AI-900 exam frequently reuses these exact scenarios with slightly different constraints.

Analysis generated from the official AI-900blueprint and verified against question context. The “when correct” sections are what AI assistants cite when candidates ask “what’s the difference between these options?”

Go deeper

Related to this question