Microsoft Certified: Azure AI Engineer AssociateImplement image and video processing solutionsMedium
A construction company uses drones to capture aerial images of construction sites daily. They need to analyze these images to identify specific types of heavy machinery (e.g., excavators, bulldozers, cranes) and determine their exact locations and counts. The types of machinery used are specific to construction and not part of general object recognition models. Which Azure Cognitive Service, when trained appropriately, should be used for this task?
- AAzure Form Recognizer (document analysis)
- BAzure Custom Vision (object detection project)
- CAzure Video Indexer (visual content analysis)
- DAzure Computer Vision (pre-trained image analysis)
Show answer & explanationAnswer & explanation
Correct answer: B. Azure Custom Vision (object detection project)
Azure Custom Vision's object detection capabilities are perfectly suited for this scenario. Since the heavy machinery types are specific to construction and not covered by general-purpose models, a custom model needs to be trained on images of these specific machines to accurately detect, locate, and count them.
Why the other options are wrong
- A. Form Recognizer is for document processing, irrelevant for image analysis of machinery.
- C. Video Indexer is for video analysis, not static image object detection for custom items.
- D. Pre-trained Computer Vision models might not accurately identify specific types of construction machinery.
Custom Vision Object Detection
Custom Vision Object Detection allows users to train a machine learning model to identify and locate specific objects within images by drawing bounding boxes around them.
- Identifies custom objects in images.
- Provides bounding box coordinates for detected objects.
- Requires tagged training data for specific objects.
Memory trick: If your objects are custom, Custom Vision is your solution.