Courseiva

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

A quality control team uses computer vision to inspect manufactured parts. They need to detect whether a part has any defects and also identify the type of defect (e.g., scratch, crack, dent) from an image. Which Azure Computer Vision capability should they use?

⚠ Common exam trap

Watch out — candidates often confuse image classification with object detection, assuming that classifying the entire image as 'defective' is sufficient, but the question explicitly requires identifying the type of each defect, which necessitates localization and multi-class output.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

B: Object detection

Object detection is the correct capability because it not only identifies the presence of defects in an image but also localizes each defect with a bounding box and classifies it into specific types (e.g., scratch, crack, dent). This meets both requirements: detecting whether a part has defects and identifying the type of each defect.

Answer analysis

Option-by-option breakdown

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

  • A: Image classification

    Why it's wrong here

    Image classification assigns a single class label to the entire image, so it cannot localize defects or distinguish multiple defect types in one scene. For a product containing both a scratch and a dent, the model would return only one label or a probability distribution, losing the spatial information needed to guide repair or rejection. It also cannot count instances, making it inadequate for quality inspection workflows that require per-defect type and location.

  • B: Object detection

    Why this is correct

    Object detection is correct because it jointly performs localization and classification by drawing bounding boxes around each defect and labeling it (e.g., 'scratch' or 'dent'). This enables quality teams to count multiple defect instances, identify their types, and record their exact coordinates for automated downstream actions. It is the standard computer vision task used in manufacturing defect inspection, including with Azure Custom Vision's object detection model.

  • C: Semantic segmentation

    Why it's wrong here

    Semantic segmentation classifies every pixel into a class, producing a precise pixel-level mask of defect regions rather than discrete bounding boxes. However, it does not separate individual instances of the same class — two adjacent scratches merge into a single mask, so counting defects is not straightforward. Training also requires expensive pixel-accurate annotations, which is overkill when simply detecting and counting defect types with a bounding box is sufficient for quality control.

  • D: Optical character recognition (OCR)

    Why it's wrong here

    Optical character recognition (OCR) is designed to extract text characters from images, such as serial numbers or part labels. It has no visual feature detectors for surface anomalies like scratches, color deviations, or structural deformations, so it cannot perform defect classification or localization. Applying OCR to visual quality inspection is a fundamental mismatch of capabilities, as OCR models are trained to recognize glyph shapes, not manufacturing flaws.

About these practice questions

One of 985 original AI-900 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

JA

Written by Johnson Ajibi, MSc IT Security

Senior Network & Security Engineer · founder of Courseiva

This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.