AI-102 Practice Test 29 — 15 Questions

Question 1

A manufacturing company uses Azure AI Custom Vision to classify defects on assembly line parts. The model is currently trained with 500 images per class and achieves 85% accuracy. The company needs to improve accuracy to 95% without collecting new images. Which THREE actions should they take?

Accepted Answer

Use data augmentation (e.g., rotation, scaling). Adjusting probability threshold can reduce false positives. Adding negative images improves model robustness. Using data augmentation generates variations from existing images. Changing domain is not applicable after training; increasing training time may help but not as directly as augmentation.

Answer

Change the domain to 'General'

Answer

Increase the number of training iterations

Question 2

Which TWO features are available in Azure Computer Vision but not in Azure Custom Vision?

Accepted Answer

Optical character recognition (OCR). A and E are correct. Azure Computer Vision provides pre-built optical character recognition (OCR) and image captioning. Custom Vision is for custom models and does not include these pre-built capabilities. B and C are wrong because Custom Vision can do image classification and object detection with custom training. D is wrong because face detection is available in Face API, not directly in Computer Vision.

Answer

Object detection

Answer

Image classification

Answer

Face detection

Question 3

You are troubleshooting an Azure AI Vision application that calls the Analyze Image API. The application suddenly returns HTTP 403 errors. The API key and endpoint have not changed. What is the most likely cause?

Accepted Answer

The API key has been regenerated or the resource is in a different region.. Option C is correct because HTTP 403 indicates the server understood the request but is refusing to authorize it. If the key and endpoint are correct, a common cause is that the API key has been regenerated or the resource has been moved to a different region. Option A is wrong because a quota exceeded returns 429, not 403. Option B is wrong because throttling returns 429. Option D is wrong because network issues may cause 5xx errors, not 403.

Answer

The image file size exceeds the maximum limit.

Answer

The service is throttling requests due to high volume.

Answer

The API call quota has been exceeded.

Question 4

Refer to the exhibit. You are using the Azure AI Face API to detect faces in an image. You need to ensure that the response includes the unique face ID for each detected face. However, the response does not contain face IDs. What is the most likely cause?

Accepted Answer

The 'detectionModel' parameter is not set; the default detection model does not return face IDs.. Option A is correct because the Face API requires that the 'detectionModel' parameter be set to 'detection_03' to return face IDs. The exhibit does not specify detectionModel, so it defaults to 'detection_01' which does not return face IDs. Option B is wrong because 'returnFaceId' is set to true. Option C is wrong because the API version is valid. Option D is wrong because face landmarks are not required for face ID.

Answer

The 'returnFaceLandmarks' parameter must be set to true.

Answer

The 'returnFaceId' parameter is set to false.

Answer

The API version is incorrect; use '2023-06-01-preview' instead.

Question 5

You are implementing a solution that uses Azure AI Vision to analyze images in a batch processing pipeline. The pipeline processes thousands of images daily. You need to minimize cost while ensuring that all images are processed within 24 hours. Which service tier should you choose?

Accepted Answer

Standard S1 tier. The Standard S1 tier provides pay-per-transaction pricing, which is cost-effective for high-volume batch processing. Free tier has limits, and S0 is too slow.

Answer

Free tier F0

Answer

Standard S0 tier

Answer

Standard S2 tier

Question 6

A company uses the Computer Vision Image Analysis API to generate captions for images. The captions are often too generic. How can they improve the descriptiveness of captions?

Accepted Answer

Use the Dense Captioning feature.. Option C is correct because using the 'dense' captioning feature provides more detailed descriptions. Option A is wrong because increasing confidence threshold reduces false positives but does not improve descriptiveness. Option B is wrong because object detection provides tags, not captions. Option D is wrong because Custom Vision can be trained for image classification, not captioning.

Answer

Use the Object Detection API instead.

Answer

Train a custom model with Custom Vision.

Answer

Increase the confidence threshold for captions.

Question 7

A company is building a computer vision solution using Azure AI Vision to analyze images of retail shelves. The solution must detect product presence and read expiration dates. Which TWO Azure AI Vision features should be used?

Accepted Answer

Object detection. Object detection identifies products on shelves. OCR reads text from images, such as expiration dates. Image captioning describes scenes, not suitable for this task. Face detection and brand detection are not relevant.

Answer

Face detection

Answer

Brand detection

Answer

Image captioning

Question 8

You call the Azure Computer Vision Analyze API with the above request body. The response includes a 'description' object with captions. Which parameter is responsible for generating captions?

Accepted Answer

Description. Option B is correct because the 'Description' visual feature generates captions. Option A is wrong because 'Categories' categorizes images. Option C is wrong because 'Tags' returns tags. Option D is wrong because 'Adult' detects adult content.

Answer

Tags

Question 9

A manufacturing company uses Azure Custom Vision to detect defects on an assembly line. The model is deployed to a container on a local edge server. Recently, the model's accuracy dropped. You suspect data drift. What should you do to monitor and retrain the model?

Accepted Answer

Periodically collect new images with labels, retrain the model in Custom Vision, and redeploy the updated container.. Option B is correct because Custom Vision does not have built-in drift detection, so you must capture new images and labels, then retrain the model in the cloud and redeploy the updated container. Option A is wrong because Custom Vision does not provide automatic retraining based on feedback. Option C is wrong because Azure Machine Learning is not directly integrated with Custom Vision for this purpose. Option D is wrong because there is no built-in drift detection; you must implement custom monitoring.

Answer

Use Azure Machine Learning data drift monitoring on the Custom Vision endpoint.

Answer

Configure Custom Vision to send alerts when drift is detected.

Answer

Enable active learning in Custom Vision to automatically retrain the model.

Question 10

You are deploying a Custom Vision model to an Azure Machine Learning managed endpoint using the above ARM template snippet. The deployment fails with a timeout error. Which parameter should you adjust?

Accepted Answer

Increase scoringTimeout to 'PT10M'. Option C is correct because the scoringTimeout of 5 minutes (PT5M) may be too short for the model to return predictions. Increasing it allows more time. Option A is wrong because authMode does not affect timeout. Option B is wrong because compute type (CPU) is not the direct cause of timeout; changing to GPU might speed up but not necessarily. Option D is wrong because the model version is not related to timeout.

Answer

Change model version to 2

Answer

Change compute to 'GPU'

Answer

Change authMode to 'AAD'

Question 11

Your application needs to determine whether two photos of the same person are of the same individual, even if they are from different angles. Which Azure AI service should you use?

Accepted Answer

Azure AI Face. Azure AI Face's 'verify' operation compares two faces and returns a confidence score for whether they belong to the same person.

Answer

Azure AI Video Indexer

Answer

Azure AI Custom Vision

Answer

Azure AI Vision OCR

Question 12

Which THREE factors should be considered when choosing between Azure AI Vision prebuilt models and Custom Vision?

Accepted Answer

Prebuilt models may not have high accuracy for industry-specific objects.. Options A, C, and E are correct. Prebuilt models have limited accuracy for niche domains (A). Custom Vision allows training on specific objects (C). Prebuilt models cover common objects, but Custom Vision is needed for custom categories (E). Option B is wrong because Custom Vision also supports images from various sources. Option D is wrong because prebuilt models do not require training data.

Answer

Custom Vision cannot process images from real-time video feeds.

Answer

Prebuilt models require a large training dataset.

Question 13

A company uses Azure AI Vision to analyze product images in an e-commerce application. The solution uses the Analyze Image API with OCR. Recently, the OCR accuracy has decreased for images with handwritten text on product labels. What should the team do to improve accuracy?

Accepted Answer

Switch to the Read API for OCR.. Option B is correct because the Read API is optimized for images with text, including handwritten text, and offers better accuracy than the Analyze Image API for OCR. Option A is wrong because Azure AI Document Intelligence (formerly Form Recognizer) is designed for structured documents. Option C is wrong because OCR is already enabled; the issue is using the wrong API. Option D is wrong because adding more training data does not apply to the prebuilt OCR model.

Answer

Enable OCR in the Analyze Image API configuration.

Answer

Use Azure AI Document Intelligence prebuilt receipt model.

Answer

Retrain the OCR model with additional handwritten samples.

Question 14

You are building a solution to automatically tag images uploaded to an Azure Storage blob container using Azure AI Vision. The solution must process images as soon as they are uploaded. Which service should you use to trigger the image analysis?

Accepted Answer

Azure Event Grid with an Azure Function trigger. Option A is correct because Azure Event Grid can trigger an Azure Function when a blob is created, which then calls Azure AI Vision. Option B is wrong because Logic Apps can also be triggered but Event Grid is more lightweight and recommended for event-driven scenarios. Option C is wrong because Azure Functions can be triggered directly by Event Grid, not by a timer. Option D is wrong because Azure Batch is for batch processing, not real-time triggers.

Answer

Azure Functions with a timer trigger

Answer

Azure Batch with a job schedule

Answer

Azure Logic Apps with a recurrence trigger

Question 15

You are designing a computer vision solution for a retail chain to detect shelf stockouts using store camera feeds. Videos are processed in near real-time. Which combination of Azure services should you use to minimize latency and cost?

Accepted Answer

Use Azure Video Analyzer for Media (formerly Video Indexer) with an Azure IoT Edge module processing video at the edge.. Option C is correct because Video Indexer is designed for video analysis but is not optimized for near real-time; Azure Video Analyzer (now part of Azure Video Indexer) with edge AI running on an IoT Edge device processes video locally, reducing latency and bandwidth costs. Option A is wrong because Video Indexer alone would stream all video to the cloud. Option B is wrong because Custom Vision is for static images, not video streams. Option D is wrong because Media Services is for encoding and streaming, not AI analysis.

Answer

Use Azure Video Indexer to analyze videos and send results to Azure SQL Database.

Answer

Use Azure Custom Vision to detect stockouts in video frames.

Answer

Use Azure Media Services to transcode video and then run Custom Vision on key frames.