Microsoft Certified: Azure AI Engineer AssociateImplement image and video processing solutionsEasy
A smart city initiative is deploying cameras at intersections to monitor traffic flow. The goal is to detect and classify different types of vehicles (cars, trucks, motorcycles, buses) in real-time video streams and count their numbers over time. The solution needs to use pre-trained models for common vehicle types. Which Azure Cognitive Service should be used?
- AAzure Face API (face detection)
- BAzure Video Indexer (visual content analysis)
- CAzure Custom Vision (object detection)
- DAzure Computer Vision (object detection)
Show answer & explanationAnswer & explanation
Correct answer: D. Azure Computer Vision (object detection)
Azure Computer Vision includes pre-trained object detection models that can identify common objects, including various types of vehicles (cars, trucks, buses, motorcycles). This service is suitable for real-time video analysis to count and classify these standard objects without requiring custom model training.
Why the other options are wrong
- A. Face API is for facial analysis, irrelevant for vehicle detection.
- B. Video Indexer is for broader video insights like transcription and topic extraction, not primarily for real-time object detection and counting.
- C. Custom Vision is for training custom models when objects are unique, but for common vehicles, a pre-trained model is more efficient.
Computer Vision Object Detection
Azure Computer Vision provides pre-trained models to detect and classify a wide range of common objects within images and video frames, returning bounding box locations and confidence scores.
- Detects common objects (e.g., vehicles, animals, people).
- Pre-trained, no custom training required for standard items.
- Provides bounding boxes and class labels.
Memory trick: For common objects on the road, Computer Vision lightens your load.