Microsoft Azure AI Fundamentals (AI-900)Describe features of computer vision workloads on AzureMedium
A smart city initiative is planning to improve urban planning by analyzing the exact boundaries of different land cover types (e.g., buildings, roads, vegetation, water bodies) from satellite imagery. This analysis requires identifying each pixel in the image and assigning it to a specific category. Which computer vision workload capability is BEST suited for this task?
- AImage Captioning
- BImage Classification
- CObject Detection
- DSemantic Segmentation
Show answer & explanationAnswer & explanation
Correct answer: D. Semantic Segmentation
Semantic Segmentation is the most appropriate capability because it classifies each pixel in an image into a predefined class, providing precise boundaries for different land cover types, which is essential for detailed urban planning and environmental analysis.
Why the other options are wrong
- A. Image Captioning generates a textual description of an image, which does not provide any visual analysis or categorization of pixels.
- B. Image Classification would assign a single label to the entire satellite image, which is insufficient for distinguishing multiple land cover types within it.
- C. Object Detection would draw bounding boxes around objects, but would not provide the exact pixel-level boundaries needed for precise land cover analysis.
Semantic Segmentation
A computer vision technique that classifies every pixel in an image into a predefined category, creating a pixel-wise mask for each object class.
- Provides precise boundaries for objects/regions.
- Used for detailed scene understanding (e.g., autonomous driving, medical imaging).
- Outputs a 'segmentation mask' where each pixel has a class label.
Memory trick: Granularity: from whole image to every pixel.