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.

• ✗

  Image Analysis (Describe image)

  Why it's wrong here

  Image Analysis (Describe image) produces a natural-language sentence or tag list summarizing the scene, such as 'a person standing in a warehouse,' but it does not output per-object bounding boxes or coordinates. A security system needs the position and size of each person to determine if they are inside a restricted zone, while a description is insufficient for spatial monitoring and alerting.

  When this WOULD be correct

  A question asking for generating a caption or description of an image, such as 'Which capability can provide a text description of the main subjects and actions in an image?'

• ✓

  Object Detection

  Why this is correct

  Object Detection identifies people by localizing each instance with a bounding box and class label, and it returns confidence scores so the system can alert when any person enters the restricted area. That fits the monitoring use case because it supports multiple object classes, can track or count people across frames, and produces coordinates that can trigger alarms.

• ✗

  Optical Character Recognition (OCR)

  Why it's wrong here

  OCR is designed to extract printed or handwritten characters from an image and returns recognized text along with text bounding boxes, but it has no object-class model for people or other physical objects. It could read signs or license plates in the area, but it cannot indicate whether a person is present, so it is fundamentally incompatible with this monitoring task.

  When this WOULD be correct

  A question asks: 'Which Azure Cognitive Services capability should be used to extract printed or handwritten text from images of documents or signs?' OCR would be the correct answer for text extraction scenarios.

• ✗

  Face Detection

  Why it's wrong here

  Face Detection identifies human faces and returns facial attributes such as landmarks, emotions, or age, but it does not perform general person detection or draw bounding boxes around entire bodies. The requirement is to detect any person present in a frame, not specifically a face; a person turned away from the camera would be missed. This option tempts because Face Detection is a well-known Computer Vision feature for locating people in images, and it would be correct if the task were to detect and frame only visible faces for identity verification or emotion analysis.

  When this WOULD be correct

  A question like 'Which Computer Vision capability should be used to detect and locate human faces in an image for a facial recognition system?' would make Face Detection correct.

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