AI-900 Practice Question: Describe features of computer vision workloads on Azure
A manufacturing company wants to use Azure AI to detect surface defects on metal parts. The team has a small set of labeled images of defective and non-defective parts, and images will be taken under various lighting conditions and angles. They need a solution that can leverage a pre-trained model and adapt it to their specific defect types with minimal new training data. Which approach should they take?
⚠ Common exam trap
Many candidates confuse the general-purpose image analysis APIs (OCR, captioning, face) with Custom Vision's specialized ability to train custom models using transfer learning, assuming any Azure AI service can be adapted to a custom task without understanding the underlying training mechanism.
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
✓
A. Use Custom Vision to train a classification or object detection model with transfer learning
Custom Vision allows you to use transfer learning, which starts from a pre-trained model and fine-tunes it on your small labeled dataset of defective and non-defective parts. This approach is ideal when you have limited training data and need to adapt the model to specific defect types under varying lighting and angles, as Custom Vision supports both classification and object detection for surface defects.
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. Use Custom Vision to train a classification or object detection model with transfer learning
Why this is correct
Custom Vision is purpose-built for custom visual inspection: it applies transfer learning from a deep convolutional neural network pre-trained on generic image datasets (e.g., ResNet variants) to your specific defect photos. You only need to provide labeled examples of surface defects and acceptable parts; the service fine-tunes the model to classify each image or localize defects with bounding boxes via object detection. This is exactly the workflow required for detecting scratches, dents, or discoloration on metal surfaces, and it can be done with a relatively small dataset.
- ✗
B. Use the Optical Character Recognition (OCR) API
Why it's wrong here
The OCR API (Read API in Azure AI Vision) is designed to extract printed or handwritten text from images using optical character recognition; it sequences characters and words, not visual surface anomalies. Surface defects like scratches, pits, or color variations have no textual representation, so the OCR service would only return text if your parts were printed with labels or part numbers. It has no concept of texture, geometry, or flaw detection, so using it to find a dent would be a categorical task mismatch.
When this WOULD be correct
A company needs to digitize printed labels on metal parts, extracting serial numbers or batch codes from images taken under varying lighting and angles.
- ✗
C. Use the Describe Image API (Image Captioning)
Why it's wrong here
The Describe Image API (also called Image Captioning within Azure AI Vision) uses a neural encoder-decoder architecture to produce a natural-language sentence summarizing the whole scene, such as 'a dark metal object on a white background.' It does not output structured classifications, confidence scores for defect categories, or bounding boxes; it merely describes overall content for use cases like alt-text generation or scene understanding. A caption like 'a scratched part' is not a deterministic defect inspection result, and the model is not trained to recognize manufacturing anomalies.
When this WOULD be correct
A company wants to automatically generate descriptive captions for product images in an e-commerce catalog to improve accessibility for visually impaired users. The Describe Image API would be correct for this image captioning task.
- ✗
D. Use the Face API
Why it's wrong here
The Face API (part of Azure AI Vision) is specialized for human faces: it detects frontal and side faces, identifies facial landmarks, and performs tasks like identity verification, attribute analysis (age, emotion), and liveness detection. The underlying models are fine-tuned on face datasets and rely on facial structure such as eyes, nose, and mouth; a metal part or plastic housing contains no face, so the API would not extract any meaningful defect signal. It is not a general-purpose visual anomaly detector and cannot be repurposed for surface inspection.
When this WOULD be correct
A company wants to verify employee identity for secure access to a manufacturing facility using a small set of labeled face images and needs a pre-trained model that can adapt with minimal data. In that case, using the Face API with transfer learning would be correct.
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.
✓A. Use Custom Vision to train a classification or object detection model with transfer learningCorrect answer▾
Why this is correct
Custom Vision is purpose-built for custom visual inspection: it applies transfer learning from a deep convolutional neural network pre-trained on generic image datasets (e.g., ResNet variants) to your specific defect photos. You only need to provide labeled examples of surface defects and acceptable parts; the service fine-tunes the model to classify each image or localize defects with bounding boxes via object detection. This is exactly the workflow required for detecting scratches, dents, or discoloration on metal surfaces, and it can be done with a relatively small dataset.
✗B. Use the Optical Character Recognition (OCR) APIWrong answer — click to see why▾
Why this is wrong here
The OCR API extracts text from images, not surface defects on metal parts, which are visual patterns unrelated to text.
★ When this WOULD be the correct answer
A company needs to digitize printed labels on metal parts, extracting serial numbers or batch codes from images taken under varying lighting and angles.
Why candidates choose this
Candidates may think OCR can detect any visual anomaly, confusing 'defect detection' with 'character recognition' due to both involving image analysis.
✗C. Use the Describe Image API (Image Captioning)Wrong answer — click to see why▾
Why this is wrong here
The Describe Image API generates captions describing the overall content of an image, but it cannot detect or classify specific surface defects on metal parts. It lacks the ability to learn new defect types from labeled training data.
★ When this WOULD be the correct answer
A company wants to automatically generate descriptive captions for product images in an e-commerce catalog to improve accessibility for visually impaired users. The Describe Image API would be correct for this image captioning task.
Why candidates choose this
Candidates may think that describing an image includes identifying defects, but the API provides general captions, not defect-specific classification or detection.
✗D. Use the Face APIWrong answer — click to see why▾
Why this is wrong here
The Face API is designed for detecting and analyzing human faces, not for detecting surface defects on metal parts. It cannot identify manufacturing defects.
★ When this WOULD be the correct answer
A company wants to verify employee identity for secure access to a manufacturing facility using a small set of labeled face images and needs a pre-trained model that can adapt with minimal data. In that case, using the Face API with transfer learning would be correct.
Why candidates choose this
Candidates might think 'defect detection' is similar to 'face detection' as both involve image analysis, leading them to mistakenly choose the Face API without considering the domain mismatch.
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
Types of AI Workloads
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
Object detection
Object detection is a computer vision technology that identifies and locates specific objects within an image or video.
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 →
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.