AWS Certified Machine Learning – Specialty — Question 325
A company that operates oil platforms uses drones to photograph locations on oil platforms that are difficult for humans to access to search for corrosion.
Experienced engineers review the photos to determine the severity of corrosion. There can be several corroded areas in a single photo. The engineers determine whether the identified corrosion needs to be fixed immediately, scheduled for future maintenance, or requires no action. The corrosion appears in an average of 0.1% of all photos.
A data science team needs to create a solution that automates the process of reviewing the photos and classifying the need for maintenance.
Which combination of steps will meet these requirements? (Choose three.)
Answer options
- A. Use an object detection algorithm to train a model to identify corrosion areas of a photo.
- B. Use Amazon Rekognition with label detection on the photos.
- C. Use a k-means clustering algorithm to train a model to classify the severity of corrosion in a photo.
- D. Use an XGBoost algorithm to train a model to classify the severity of corrosion in a photo.
- E. Perform image augmentation on photos that contain corrosion.
- F. Perform image augmentation on photos that do not contain corrosion.
Correct answer: A, D, E
Explanation
An object detection algorithm (Option A) is necessary to locate and isolate multiple instances of corrosion within a single image. Because corrosion only appears in 0.1% of the images, applying image augmentation to the corrosion-positive images (Option E) is critical to address the extreme class imbalance. Finally, an XGBoost classifier (Option D) can be trained on tabular features extracted from the detected corrosion regions to accurately categorize the severity of the maintenance required.