Microsoft Azure AI Fundamentals (AI-900)Describe features of computer vision workloads on AzureEasy
A company is developing a system to monitor a large industrial facility for security breaches. They need to identify when unauthorized individuals enter specific restricted zones, trigger an alert, and record the event. Which computer vision workload capability is MOST suitable for this scenario?
- AObject Detection
- BOptical Character Recognition (OCR)
- CImage Classification
- DImage Captioning
Show answer & explanationAnswer & explanation
Correct answer: A. Object Detection
Object Detection allows the system to identify and locate specific objects (like people) within an image or video frame, making it ideal for monitoring restricted zones for unauthorized entry. It provides bounding box coordinates for each detected object.
Why the other options are wrong
- B. OCR extracts text from images, which is not relevant for identifying unauthorized individuals.
- C. Image Classification identifies the primary subject of an entire image, not specific objects within it or their location.
- D. Image Captioning generates a descriptive text summary of an image, but does not pinpoint specific objects or their locations for security monitoring.
Object Detection
A computer vision technique that identifies and locates instances of objects within an image or video, drawing bounding boxes around each detected object.
- Identifies specific objects (e.g., people, cars, animals).
- Provides location of each object (bounding box).
- Used for counting, tracking, and security monitoring.
Memory trick: Vision sees, detects, and describes.