Week 5 Image Segmentation and Registration Flashcards
What does image segmentation do?
Image segmentation divides an image into regions with similar properties.
What is thresholding in segmentation?
Thresholding is a basic technique for segmentation based on intensity values.
What is Otsu’s method used for?
Otsu’s method is an automatic thresholding technique for bimodal histograms.
What is region-based segmentation?
Region-based segmentation groups pixels with similar properties, such as intensity or colour.
What does the Watershed algorithm use for segmentation?
The Watershed algorithm uses gradient information for segmentation.
What is the purpose of K-means clustering in segmentation?
K-means clustering is used for unsupervised segmentation based on pixel features.
What does semantic segmentation achieve?
Semantic segmentation assigns a class label to every pixel in an image.
What is image registration used for?
Image registration aligns two or more images to a common coordinate system.
What does RANSAC do in image registration?
RANSAC (Random Sample Consensus) is often used for robust transformation estimation.
What are preprocessing steps for improving registration accuracy?
Preprocessing steps, like denoising and contrast enhancement, improve registration accuracy.
What is instance segmentation?
Instance segmentation identifies and separates individual objects within the same class.
What modern methods are used for segmentation?
Modern methods include deep learning-based approaches like Fully Convolutional Networks (FCNs).
What is U-Net used for in segmentation?
U-Net is a popular architecture for medical image segmentation.
Why is image segmentation critical?
Image segmentation is critical for applications like medical imaging, autonomous driving, and object tracking.
What are challenges in image segmentation?
Challenges include handling occlusions. Noise. And computational complexity.
