AI-900 Practice Question: Describe features of computer vision workloads on Azure
A wildlife research team uses drone imagery to monitor penguin populations in a remote area. The penguins are small, blend into the rocky background, and are often only partially visible. The team has a limited set of 500 labeled drone images showing penguins. They want to build a system that accurately detects and counts penguins. Which approach should they take using Azure AI services?
⚠ Common exam trap
Test-takers frequently confuse image classification with object detection, assuming a single label per image can solve a counting problem, or overestimate the generic API's ability to handle niche, low-contrast objects without custom training.
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
✓
Train a Custom Vision object detection model using the labeled images.
The pre-built Computer Vision object detection API is optimized for common objects and may not perform well on small, camouflaged penguins in rocky terrain. Custom Vision allows the team to train a dedicated object detection model using their 500 labeled images, enabling the model to learn the specific visual features of penguins in this challenging environment. This approach is ideal for domain-specific detection tasks where off-the-shelf models lack accuracy.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Use the pre-built Computer Vision object detection API directly.
Why it's wrong here
The pre-built Computer Vision object detection API is a fixed, general-purpose model trained on common object categories such as people, vehicles, and household animals; it does not include a penguin class and cannot be retrained or fine-tuned by users. Attempting to use it on drone imagery would likely yield no detections or spurious low-confidence boxes for penguins, because the model has no learned visual representation of that specific species. Thus, a custom object detection model is required to recognize and accurately count penguins.
When this WOULD be correct
A question where the objects are common (e.g., cars, dogs) and the user has no labeled data, requiring a zero-shot solution. For example: 'A company needs to detect common office items in photos without any labeled training data.'
- ✓
Train a Custom Vision object detection model using the labeled images.
Why this is correct
Custom Vision enables training a specialized object detection model with a small set of labeled images. With only one object class ('penguin'), 500 images are more than sufficient to achieve good accuracy for detection and counting.
- ✗
Use the Computer Vision Image Analysis API with the 'dense captioning' feature.
Why it's wrong here
Dense captioning in the Computer Vision Image Analysis API annotates image regions with natural-language descriptions (e.g., "a bird in the snow") and coarse bounding boxes, but it does not return class-verified detections with confidence scores suitable for counting. Its underlying model is trained for general scene understanding and lacks a penguin class, so it may miss, merge, or misdescribe individual birds. Therefore, while it can hint at content, it is not an object detection or counting mechanism.
- ✗
Train a Custom Vision image classification model with the labeled images.
Why it's wrong here
Custom Vision image classification trains a model to assign one global label to the entire input image, such as "penguin" or "no penguin," but it produces no bounding boxes or localization information. Even with accurately labeled penguin images, the model cannot determine where each penguin is or how many distinct individuals occupy the frame. Counting individuals requires an object detection model outputting separate bounding boxes and per-instance confidences, making classification fundamentally insufficient for this task.
When this WOULD be correct
If the team only needed to determine whether penguins are present in an image (e.g., presence/absence monitoring) and did not require counting or localization, training a Custom Vision image classification model would be appropriate.
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.
✓Train a Custom Vision object detection model using the labeled images.Correct answer▾
Why this is correct
Custom Vision enables training a specialized object detection model with a small set of labeled images. With only one object class ('penguin'), 500 images are more than sufficient to achieve good accuracy for detection and counting.
✗Use the pre-built Computer Vision object detection API directly.Wrong answer — click to see why▾
Why this is wrong here
The pre-built Computer Vision object detection API is not trained on penguins in rocky terrain and would fail to accurately detect partially visible, small penguins blending into the background, especially with only 500 labeled images for fine-tuning.
★ When this WOULD be the correct answer
A question where the objects are common (e.g., cars, dogs) and the user has no labeled data, requiring a zero-shot solution. For example: 'A company needs to detect common office items in photos without any labeled training data.'
Why candidates choose this
Candidates may assume pre-built APIs work for any object detection task without considering the need for domain-specific training on rare or camouflaged objects.
✗Train a Custom Vision image classification model with the labeled images.Wrong answer — click to see why▾
Why this is wrong here
Image classification assigns a single label to the entire image, not detecting or localizing individual objects. Since the team needs to count penguins that are small and partially visible, object detection is required to identify each penguin's location.
★ When this WOULD be the correct answer
If the team only needed to determine whether penguins are present in an image (e.g., presence/absence monitoring) and did not require counting or localization, training a Custom Vision image classification model would be appropriate.
Why candidates choose this
Candidates may confuse image classification with object detection, assuming that classification can also count objects, or they may not fully understand the difference between the two tasks.
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
Learn chapter
Azure Machine Learning Studio
Key term
Computer vision
Computer vision is a field of artificial intelligence that enables computers to interpret and make decisions based on visual data from the world, such as images and videos.
Key term
Model
In IT and AI, a model is a trained mathematical representation that learns patterns from data to make predictions or decisions.
About these practice questions
Courseiva writes every AI-900 question from scratch — 985 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or 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.