AI-900 Practice Question: Describe features of computer vision workloads on Azure
A retail chain uses ceiling-mounted cameras to monitor shelf inventory. They need to identify and locate individual products (e.g., a specific brand of cereal) within an image and count how many are present. Which Azure Computer Vision capability should they use?
⚠ Common exam trap
Many candidates confuse object detection with image classification, assuming that labeling the image as 'cereal' is sufficient to count items, when in fact object detection is required for instance-level localization and counting.
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
✓
Object detection
Object detection is the correct capability because it not only identifies the presence of a specific product (e.g., a brand of cereal) within an image but also localizes each instance by drawing bounding boxes around them, enabling an accurate count. Image classification would only label the entire image as containing cereal without locating individual boxes, while OCR and semantic segmentation serve different purposes (text extraction and pixel-level labeling, respectively).
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 classification
Why it's wrong here
Image classification treats an entire image as a single unit and assigns one categorical label—such as 'cereal aisle'—based on the predominant visual content. It uses global feature extraction (e.g., via convolutional neural networks with global average pooling) and produces no bounding boxes, spatial coordinates, or instance counts. Because it collapses all objects into one label, it cannot tell you how many distinct products are present or where they are located, so it is unsuitable for inventory monitoring.
- ✓
Object detection
Why this is correct
Object detection is the correct choice because it jointly performs localization and classification: for each object of interest, it returns a bounding box (x, y, width, height) and a class label, along with a confidence score. Modern detectors like Faster R-CNN use region proposal networks, while single-shot detectors like YOLO predict boxes and classes directly from feature maps. The presence of per-instance boxes lets the retail chain count every product in the camera's field of view, which is exactly what the monitoring scenario requires.
- ✗
Optical character recognition (OCR)
Why it's wrong here
Optical character recognition (OCR) is designed exclusively for extracting textual content—printed or handwritten characters—from images. It typically involves text detection to localize text regions and then recognition via recurrent networks or attention-based decoders to transcribe the characters. Products in a retail aisle are non-text objects; even if packaging contains logos or labels, OCR will only read the embedded text and cannot generate bounding boxes around the product itself, nor can it count individual items, so it is irrelevant to this use case.
- ✗
Semantic segmentation
Why it's wrong here
Semantic segmentation assigns a class label to every pixel in the image, creating a dense pixel-level map (e.g., 'shelf', 'product', 'floor') rather than discrete object instances. Because it classifies pixels but does not differentiate between separate objects of the same class—for example, two identical cereal boxes touching each other would be labeled as one continuous 'product' region—it cannot enumerate individual products reliably. Additionally, predicting a mask for every pixel is computationally heavier than object detection and is overkill for a task that only needs to locate and count items.
Go deeper
Related to this question
Learn chapter
Regression and Classification
Key term
Classification
Classification is a supervised machine learning technique used to predict a category or class label for new data based on patterns learned from labeled training data.
Key term
Inventory
Inventory is the complete list of hardware, software, and network assets that an organization owns or manages, tracked for maintenance, security, and budgeting purposes.
About these practice questions
This AI-900 question is part of Courseiva's 985-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
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.