Microsoft Azure AI Fundamentals (AI-900)Describe features of computer vision workloads on AzureMedium
A researcher is studying wildlife populations and needs to count specific animal species in aerial photographs. The system should not only count the animals but also indicate the general area where each animal is located within the image. Which computer vision capability is most appropriate for this task?
- AObject Detection
- BOptical Character Recognition (OCR)
- CImage Captioning
- DImage Classification
Show answer & explanationAnswer & explanation
Correct answer: A. Object Detection
Object Detection is the most appropriate capability as it can identify and locate individual instances of specific animal species within an image by drawing bounding boxes around them, allowing for both counting and general area localization.
Why the other options are wrong
- B. OCR is for extracting text, which is irrelevant to counting animals.
- C. Image Captioning generates a textual description of the image, not a count or location of specific objects.
- D. Image Classification would only tell if an animal species is present in the image, not how many or where they are.
Object Detection for Counting
Using object detection models to identify and localize individual instances of objects within an image, thereby enabling accurate counting of those objects.
- Each detected object gets a bounding box and a class label.
- Useful for inventory, traffic analysis, and wildlife monitoring.
- Provides both 'what' and 'where' information.
Memory trick: Detect objects to count and place them.