AI-102 Implement computer vision solutions Practice Question
You are training an Azure Custom Vision object detection model to locate pallets in warehouse photos. Your training set contains 500 images, but only 40 images include pallets, while the rest are empty aisles. The model performs poorly, often missing pallets. You need to improve detection while keeping training time reasonable. What should you do first?
⚠ Common exam trap
The trap here is reaching for a threshold or domain tweak, when the real cause is too few labeled positive examples relative to empty background images.
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
✓
Add more labeled images that contain pallets in varied conditions and balance the dataset.
Object detection models learn from the distribution of labeled examples. When pallet images are only 8 percent of the set, the model is biased toward the dominant empty-aisle class and misses pallets. Adding more varied, labeled pallet images and balancing classes improves the positive signal. Threshold tuning, domain changes, or simple retraining do not correct the underlying data imbalance.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Enable the 'General' domain and retrain with the same dataset without changes.
Why it's wrong here
Retraining on the same imbalanced data with a different domain will reproduce the same bias toward empty aisles. Domain choice affects feature extraction style, not the ratio of positive to negative samples. Without adding representative pallet images, the model still lacks enough signal to detect pallets reliably.
- ✗
Switch the project domain to a compact domain to speed up training.
Why it's wrong here
Compact domains are optimized for exporting models to constrained devices and generally trade some accuracy for size. Changing the domain does not fix the class imbalance that causes missed detections and may reduce accuracy further. Speed is not the primary issue; detection quality is, so this does not address the scenario.
- ✗
Increase the probability threshold so fewer false positives are returned.
Why it's wrong here
Raising the probability threshold filters low-confidence predictions at inference time; it does not teach the model to find pallets it currently misses. In fact, a higher threshold would suppress even more detections, worsening the recall problem. Threshold tuning addresses precision, not the underlying class imbalance causing missed pallets.
- ✓
Add more labeled images that contain pallets in varied conditions and balance the dataset.
Why this is correct
The model misses pallets because positive examples are scarce relative to empty aisles, so it learns the background class too strongly. Adding and labeling more pallet images across lighting, angles, and stacking arrangements gives the model representative positive features, improving recall. Balancing the dataset directly targets the root cause rather than masking it with threshold changes.
Go deeper
Related to this question
About these practice questions
This AI-102 question is part of Courseiva's 761-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 and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Microsoft exam blueprint
This AI-102 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-102 exam.