1. Who is responsible for grouping similar pixels in image clustering?
(A) Feature extractor
(B) Clustering algorithm
(C) Noise reducer
(D) Histogram equalizer
2. Which clustering technique does not require the number of clusters to be predefined?
(A) K-means
(B) Mean shift
(C) Hierarchical clustering
(D) DBSCAN
3. What does K represent in K-means clustering?
(A) The number of features
(B) The number of data points
(C) The number of clusters
(D) The number of iterations
4. Which distance metric is commonly used in image clustering?
(A) Cosine similarity
(B) Hamming distance
(C) Euclidean distance
(D) Manhattan distance
5. In clustering, an image is treated as a collection of what?
(A) Color intensities
(B) Feature descriptors
(C) Pixels
(D) Data points
6. Which method builds a hierarchy of clusters?
(A) K-means
(B) Fuzzy C-means
(C) Hierarchical clustering
(D) DBSCAN
7. What is a key limitation of K-means clustering?
(A) Requires labeled data
(B) Sensitive to noise and outliers
(C) Only works with grayscale images
(D) Requires kernel functions
8. In image clustering, what role do color histograms play?
(A) Segmentation
(B) Feature representation
(C) Noise removal
(D) Interpolation
9. Which of the following is a soft clustering algorithm?
(A) K-means
(B) Agglomerative clustering
(C) DBSCAN
(D) Fuzzy C-means
10. What does DBSCAN stand for?
(A) Density-Based Spatial Clustering of Applications with Noise
(B) Direct Binary Spatial Clustering and Normalization
(C) Discrete Binary Scan Algorithm
(D) Density-Based Segmentation Clustering Algorithm
11. What is the main advantage of hierarchical clustering?
(A) It scales well to large datasets
(B) It requires fewer computations
(C) It reveals the structure in the data
(D) It provides real-time segmentation
12. Which clustering method involves merging or splitting clusters iteratively?
(A) DBSCAN
(B) Hierarchical clustering
(C) K-means
(D) GMM
13. Which type of clustering allows each data point to belong to multiple clusters?
(A) Hard clustering
(B) K-means
(C) Fuzzy clustering
(D) Agglomerative clustering
14. Which algorithm uses centroids to define clusters?
(A) DBSCAN
(B) Spectral clustering
(C) K-means
(D) OPTICS
15. Which of the following clustering algorithms works best for arbitrary-shaped clusters?
(A) K-means
(B) DBSCAN
(C) K-medoids
(D) Fuzzy C-means
16. What is the purpose of silhouette score in clustering?
(A) Evaluate cluster quality
(B) Reduce image noise
(C) Detect outliers
(D) Compute centroids
17. Which clustering method is deterministic and not random?
(A) K-means
(B) DBSCAN
(C) Agglomerative
(D) Spectral
18. What is over-segmentation in image clustering?
(A) Ignoring edges
(B) Creating too many small regions
(C) Combining all pixels into one cluster
(D) Mislabeling cluster centroids
19. In clustering, what does a dendrogram represent?
(A) Cluster centers
(B) Cluster hierarchy
(C) Feature distribution
(D) Segmentation boundaries
20. What is a drawback of DBSCAN?
(A) Cannot handle noise
(B) Does not detect small clusters
(C) Requires specifying epsilon
(D) Always needs a distance matrix
21. Which clustering method is most appropriate for non-convex shapes?
(A) K-means
(B) Hierarchical
(C) DBSCAN
(D) PCA
22. What does clustering in image segmentation primarily aim to achieve?
(A) Object detection
(B) Color enhancement
(C) Grouping similar regions
(D) Filtering noise
23. What is the role of the ‘elbow method’ in K-means clustering?
(A) Initializing centroids
(B) Determining optimal K
(C) Calculating distance matrix
(D) Merging clusters
24. Which algorithm merges clusters based on a linkage criterion?
(A) K-means
(B) DBSCAN
(C) Mean Shift
(D) Hierarchical clustering
25. Why is normalization important before clustering images?
(A) Reduces resolution
(B) Enhances contrast
(C) Scales features equally
(D) Removes color
26. Which clustering method partitions data into k groups and minimizes intra-cluster variance?
(A) DBSCAN
(B) Spectral
(C) K-means
(D) Agglomerative
27. Which clustering method is sensitive to initialization?
(A) Hierarchical
(B) DBSCAN
(C) K-means
(D) Fuzzy C-means
28. Which clustering method works by finding modes in the feature space?
(A) Mean Shift
(B) K-means
(C) DBSCAN
(D) OPTICS
29. Which type of clustering does not assume prior knowledge of data distribution?
(A) Gaussian Mixture Models
(B) DBSCAN
(C) Spectral clustering
(D) PCA
30. What is the goal of spectral clustering?
(A) Estimate Gaussian parameters
(B) Compute mean shift
(C) Partition graph-based similarity
(D) Reduce color dimensions
31. What makes Fuzzy C-means different from K-means?
(A) Uses centroids
(B) Hard assignment of data points
(C) Allows soft assignments
(D) Faster computation
32. Which clustering technique uses eigenvalues of the similarity matrix?
(A) DBSCAN
(B) K-means
(C) Spectral clustering
(D) Agglomerative clustering
33. What does high intra-cluster similarity indicate?
(A) Poor segmentation
(B) Well-defined clusters
(C) Noise in the data
(D) Irregular pixel spacing
34. Which technique visualizes cluster tendency before applying clustering?
(A) PCA
(B) Heatmap
(C) Dendrogram
(D) VAT (Visual Assessment of cluster Tendency)
35. Why is image clustering considered an unsupervised technique?
(A) It requires labels
(B) It uses predefined classes
(C) It does not use labeled data
(D) It performs feature extraction
36. What type of features are typically used in clustering color images?
(A) Histogram of gradients
(B) RGB values
(C) Optical flow
(D) Edge maps
37. Which algorithm is efficient for clustering large datasets with noise?
(A) K-means
(B) DBSCAN
(C) Hierarchical
(D) K-medoids
38. What is the main role of a centroid in clustering?
(A) Define edges
(B) Measure similarity
(C) Represent the center of a cluster
(D) Segment noise
39. Which method assumes data comes from a mixture of Gaussians?
(A) K-means
(B) DBSCAN
(C) GMM
(D) Spectral clustering
40. In image clustering, what does feature extraction help with?
(A) Removing clusters
(B) Reducing image brightness
(C) Representing images in numerical form
(D) Compressing images
41. Which clustering algorithm is based on density estimation?
(A) DBSCAN
(B) K-means
(C) Agglomerative
(D) PCA
42. What makes clustering different from classification?
(A) Requires labels
(B) Unsupervised nature
(C) Predicts exact class
(D) Uses training data
43. Which of the following is not a clustering algorithm?
(A) GMM
(B) K-means
(C) PCA
(D) DBSCAN
44. Which term defines how compact and well-separated clusters are?
(A) Compactness Index
(B) Distance Matrix
(C) Clustering Score
(D) Silhouette Coefficient
45. In image clustering, what is the result of under-clustering?
(A) Too many regions
(B) Merged distinct regions
(C) Missing image edges
(D) Noisy segmentation
46. Which of these is NOT typically a step in image clustering?
(A) Feature extraction
(B) Label training
(C) Cluster assignment
(D) Distance computation
47. What is the outcome of clustering applied to image datasets?
(A) Supervised labels
(B) Pixel-wise classification
(C) Grouped similar images or regions
(D) Color balancing
48. Which algorithm handles clusters of varying densities well?
(A) K-means
(B) DBSCAN
(C) Agglomerative clustering
(D) Fuzzy C-means
49. What is the purpose of applying clustering in content-based image retrieval?
(A) Increase resolution
(B) Reduce noise
(C) Group similar images
(D) Detect edges
50. Which clustering technique is most sensitive to the initial choice of cluster centers?
(A) DBSCAN
(B) K-means
(C) Agglomerative clustering
(D) Mean Shift
