1. Which of the following is a fundamental requirement for region splitting and merging?
(A) Image must be binary
(B) Homogeneity criterion
(C) Image must be color
(D) Frequency domain conversion
2. What is the main goal of region splitting in image segmentation?
(A) Combine similar regions
(B) Apply morphological operations
(C) Divide image into smaller homogeneous parts
(D) Convert color space
3. Which of the following methods can be considered a top-down approach?
(A) Region merging
(B) Region splitting
(C) Thresholding
(D) K-means clustering
4. Which technique starts with the whole image and recursively divides it?
(A) Region merging
(B) Region growing
(C) Region splitting
(D) Watershed
5. In region merging, two regions are merged if they satisfy which condition?
(A) Same intensity range
(B) Homogeneity condition
(C) Different histograms
(D) Maximum entropy
6. What does the region splitting and merging technique combine?
(A) Frequency and spatial domain
(B) Thresholding and edge detection
(C) Top-down and bottom-up strategies
(D) Morphological and transform domain methods
7. What is the minimum region size usually defined for splitting to stop?
(A) Depends on image format
(B) Based on filter size
(C) User-defined or predefined threshold
(D) Always 1 pixel
8. Which region-based segmentation method is more sensitive to noise?
(A) Region growing
(B) Region splitting
(C) Watershed
(D) K-means clustering
9. What is the smallest unit into which an image is divided in region splitting?
(A) Block
(B) Pixel
(C) Quad region
(D) Superpixel
10. Which of the following can cause over-segmentation in region splitting?
(A) Large block size
(B) Loose homogeneity criterion
(C) Strict homogeneity criterion
(D) Region merging
11. What data structure is commonly used in region splitting?
(A) Stack
(B) Graph
(C) Quad-tree
(D) Matrix chain
12. Which of the following statements is true about region merging?
(A) It starts with the full image
(B) It splits the image into four equal parts
(C) It merges adjacent similar regions
(D) It cannot be reversed
13. Region merging can be considered a:
(A) Top-down process
(B) Bottom-up process
(C) Linear process
(D) Thresholding technique
14. Which of these criteria is not typically used in defining homogeneity?
(A) Intensity
(B) Color
(C) Texture
(D) Edge gradient
15. What is a common problem in both region splitting and merging?
(A) Quantization
(B) Choosing an appropriate homogeneity criterion
(C) Color conversion
(D) Histogram equalization
16. Why is region splitting and merging considered recursive?
(A) Because it uses recursion in implementation
(B) It repeats operations until a condition is met
(C) It loops back to initial pixel values
(D) It always requires Fourier transform
17. What is the initial step in region splitting?
(A) Divide image into small squares
(B) Apply homogeneity check
(C) Start with the whole image
(D) Apply thresholding
18. What typically triggers a split in region splitting?
(A) Edge detection
(B) Homogeneity test fails
(C) Brightness exceeds 128
(D) Histogram shifts
19. In a quad-tree representation, what does each node represent?
(A) Edge
(B) Region
(C) Pixel
(D) Histogram
20. What is the termination condition in region merging?
(A) All pixels are merged
(B) No two adjacent regions satisfy homogeneity
(C) Regions have same shape
(D) The histogram becomes uniform
21. Which is an advantage of using region splitting and merging?
(A) Fully automatic segmentation
(B) Less computation
(C) Handles noise better than edge-based methods
(D) Always gives a single region
22. Which segmentation approach is more sensitive to seed location?
(A) Region splitting
(B) Region merging
(C) Region growing
(D) Quad-tree
23. What happens when the homogeneity criterion is too strict?
(A) Under-segmentation
(B) Noise is reduced
(C) Over-segmentation
(D) Smoothing is enhanced
24. What happens when the homogeneity criterion is too loose?
(A) More details are captured
(B) Under-segmentation
(C) Over-segmentation
(D) Edge details improve
25. Which of the following can not be used as a homogeneity criterion?
(A) Color variance
(B) Intensity range
(C) Edge sharpness
(D) Texture uniformity
26. How many subregions are created in each split using quad-tree?
(A) 2
(B) 3
(C) 4
(D) 8
27. Which operation is applied after region splitting to avoid over-segmentation?
(A) Filtering
(B) Region merging
(C) Histogram equalization
(D) Fourier transform
28. What must be checked before merging two regions?
(A) Location
(B) Shape
(C) Homogeneity
(D) Gradient
29. Region splitting and merging algorithms work on which domain?
(A) Frequency
(B) Spatial
(C) Time
(D) Compressed
30. What is an application of region splitting and merging?
(A) Image compression
(B) Noise filtering
(C) Medical image segmentation
(D) Image encryption
31. Which approach is more computationally expensive?
(A) Splitting
(B) Merging
(C) Combined splitting and merging
(D) Thresholding
32. What is the major limitation of region splitting alone?
(A) Slow processing
(B) Cannot detect edges
(C) Over-segmentation
(D) Under-segmentation
33. Which type of image is best suited for region splitting and merging?
(A) Binary
(B) High-contrast images
(C) Homogeneous regions
(D) Noisy images
34. Region splitting and merging methods are most suitable when:
(A) Texture variation is high
(B) Regions have clear boundaries
(C) Edges are strong
(D) Homogeneity is consistent
35. Why is post-processing often applied after splitting and merging?
(A) To enhance colors
(B) To apply filtering
(C) To remove small noisy regions
(D) To smooth gradients
36. Which of the following algorithms is based on region merging?
(A) Mean-shift
(B) Canny edge detector
(C) Otsu’s method
(D) Sobel operator
37. In what order are operations performed in combined splitting and merging?
(A) Split → Merge
(B) Merge → Split
(C) Random
(D) Merge only
38. What is the output of region splitting and merging segmentation?
(A) Smoothed image
(B) Edge map
(C) Labelled regions
(D) Binary mask
39. Which of the following tools are used to visualize region splits?
(A) Histogram
(B) Quad-tree
(C) Edge map
(D) Heat map
40. When applying merging, adjacent regions are compared for:
(A) Sharpness
(B) Homogeneity
(C) Texture
(D) Orientation
41. Which method works best for images with large uniform regions?
(A) Region growing
(B) Edge detection
(C) Region splitting and merging
(D) Histogram equalization
42. What is typically required after region merging to refine boundaries?
(A) Edge detection
(B) Filtering
(C) Region growing
(D) Watershed
43. What technique improves segmentation quality after splitting and merging?
(A) Histogram stretching
(B) Smoothing filter
(C) Morphological processing
(D) Thresholding
44. Which of the following improves performance of splitting and merging?
(A) Adaptive homogeneity threshold
(B) Global histogram equalization
(C) Frequency transform
(D) Color space change
45. Which segmentation method uses both merging and splitting adaptively?
(A) Region growing
(B) Watershed
(C) Adaptive region splitting and merging
(D) Canny edge detection
46. Which mathematical tool can represent splitting structure?
(A) Graph
(B) Tree
(C) Quad-tree
(D) Matrix
47. What happens if the image is noisy and no preprocessing is done?
(A) Better segmentation
(B) Fewer merges
(C) Over-segmentation
(D) No effect
48. When are two regions usually merged in quad-tree based methods?
(A) If they differ significantly
(B) If they are not neighbors
(C) If they satisfy homogeneity
(D) If histogram is same
49. Which of the following is most useful for real-time segmentation?
(A) Otsu’s method
(B) Region splitting and merging
(C) Canny edge detection
(D) Watershed transform
50. What is one key challenge in implementing region splitting and merging?
(A) Edge detection accuracy
(B) Defining global threshold
(C) Choosing efficient homogeneity criteria
(D) Histogram smoothing
51. Which of the following is not required in region merging?
(A) Similarity check
(B) Spatial adjacency
(C) Color histogram
(D) Region overlap
52. What is the result when the homogeneity threshold is zero?
(A) All regions are merged
(B) No region is split
(C) Each pixel becomes a region
(D) Maximum compression
