AI-102 Plan and manage an Azure AI solution Practice Question
A company is using Azure AI Vision to analyze images from a manufacturing line. The solution must detect defects in real-time. The team discovers that the model's accuracy drops significantly when images are captured under different lighting conditions. What is the best approach to improve the model's robustness?
⚠ Common exam trap
The trap is choosing inference-time preprocessing (normalize lighting) as a quick fix instead of addressing the training data distribution, which is the actual cause of the robustness gap.
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
✓
Retrain the model using images captured under various lighting conditions, using data augmentation.
The accuracy drop is caused by a domain shift: the model was trained on images with limited lighting variation but is deployed under diverse lighting. Retraining with images captured under various lighting conditions and applying data augmentation (brightness, contrast, exposure jitter) teaches the model to be invariant to those variations, directly improving robustness. This addresses the root cause rather than masking it at inference time.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Apply image pre-processing to normalize lighting before sending to the model.
Why it's wrong here
Normalising lighting alters pixel values before inference but cannot teach the model to recognise defects under varied illumination, so accuracy still degrades. Pre-processing suits fixed, predictable capture conditions; the scenario needs augmentation or retraining with images spanning the actual lighting range.
- ✗
Increase the number of training images without varying lighting conditions.
Why it's wrong here
Adding more images under the same lighting reinforces the existing bias, so accuracy still collapses under different conditions. It is tempting because more data usually helps, but robustness requires training images that vary lighting, since the model must learn the defect features independent of illumination.
- ✓
Retrain the model using images captured under various lighting conditions, using data augmentation.
Why this is correct
Accuracy drops because the model has not learned invariance to illumination, a covariate shift between training and inference data. Retraining with images spanning varied lighting, plus augmentation that synthesises those variations, exposes the model to the real-world distribution it must handle, directly restoring robustness under the differing lighting conditions the stem describes.
- ✗
Use a pre-built model from Azure AI Vision instead of a custom model.
Why it's wrong here
Pre-built models recognise general categories such as objects or faces, not manufacturing defects, so they cannot match a custom defect classifier's accuracy. It is tempting because pre-built models need no training, but robustness to lighting variation comes from training a custom model on images captured under those varied conditions.
Go deeper
Related to this question
About these practice questions
Courseiva writes every AI-102 question from scratch — 761 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 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.