1. Which edge detection operator uses a pair of 3×3 convolution masks to detect horizontal and vertical edges?
(A) Roberts
(B) Sobel
(C) LoG
(D) Canny
2. Which edge detection method combines Gaussian smoothing with Laplacian operation?
(A) Prewitt
(B) Roberts
(C) LoG
(D) Canny
3. Which operator is best for detecting diagonal edges in an image?
(A) Sobel
(B) Roberts
(C) Prewitt
(D) LoG
4. Which method is known for using non-maximum suppression in its edge detection process?
(A) Sobel
(B) Canny
(C) Prewitt
(D) Roberts
5. Which edge detector applies double thresholding to identify potential edges?
(A) Canny
(B) Sobel
(C) Prewitt
(D) LoG
6. Which of the following edge detectors performs best in the presence of noise?
(A) Roberts
(B) Sobel
(C) Canny
(D) Prewitt
7. What is the size of the Roberts operator masks?
(A) 3×3
(B) 2×2
(C) 5×5
(D) 1×2
8. Which method is a derivative-based edge detector that uses two kernels to approximate gradients?
(A) Sobel
(B) Laplacian
(C) Canny
(D) LoG
9. Which edge detection operator uses the square root of the sum of squared gradients?
(A) Canny
(B) Prewitt
(C) Sobel
(D) Roberts
10. Which operator is considered the simplest in terms of implementation?
(A) Canny
(B) LoG
(C) Roberts
(D) Prewitt
11. Which step in the Canny algorithm ensures only the most prominent edges remain?
(A) Gaussian Smoothing
(B) Gradient Calculation
(C) Non-Maximum Suppression
(D) Double Thresholding
12. Which edge detection technique uses the second derivative?
(A) LoG
(B) Sobel
(C) Prewitt
(D) Canny
13. In which operator are the gradient directions computed using arctangent?
(A) Sobel
(B) Canny
(C) Roberts
(D) Prewitt
14. Which operator uses a 3×3 kernel with equal weights for vertical and horizontal gradients?
(A) Sobel
(B) Prewitt
(C) Roberts
(D) LoG
15. Which operator combines edge detection with smoothing in a single step?
(A) Sobel
(B) Canny
(C) LoG
(D) Prewitt
16. Which operator is more sensitive to noise due to its small kernel size?
(A) Prewitt
(B) Sobel
(C) Roberts
(D) Canny
17. Which edge detection operator is directional and not rotationally symmetric?
(A) Canny
(B) Sobel
(C) LoG
(D) Laplacian
18. Which of the following is NOT a step in the Canny edge detection process?
(A) Gradient calculation
(B) Non-maximum suppression
(C) Morphological dilation
(D) Hysteresis thresholding
19. Which edge detection method gives thick edges and is not ideal for accurate localization?
(A) LoG
(B) Canny
(C) Prewitt
(D) Roberts
20. Which method is ideal for detecting edges with low contrast but good noise suppression?
(A) Sobel
(B) Prewitt
(C) Canny
(D) Roberts
21. Which operator is a discrete differentiation operator?
(A) Canny
(B) LoG
(C) Sobel
(D) Fourier
22. What is the primary limitation of the Roberts operator?
(A) Too complex
(B) Poor localization
(C) Sensitive to noise
(D) Requires large kernel
23. Which of the following does Canny use to connect edge pixels?
(A) Morphological closing
(B) Gradient linking
(C) Edge tracking by hysteresis
(D) Median filtering
24. Which of these uses the Laplacian operator after Gaussian filtering?
(A) Sobel
(B) LoG
(C) Roberts
(D) Canny
25. Which method provides better edge continuity?
(A) Prewitt
(B) Sobel
(C) Canny
(D) Roberts
26. Which operator approximates the gradient magnitude?
(A) Sobel
(B) Fourier
(C) Laplacian
(D) LoG
27. In LoG, what shape is used to perform convolution?
(A) Circle
(B) Gaussian
(C) Square
(D) Triangle
28. Which edge detector typically produces the thinnest edges?
(A) Roberts
(B) Sobel
(C) Canny
(D) Prewitt
29. Which operator performs better at corners and curves?
(A) LoG
(B) Canny
(C) Prewitt
(D) Roberts
30. Which operator is symmetric and uses weights based on pixel distance?
(A) Sobel
(B) Prewitt
(C) LoG
(D) Laplacian
31. Which operator approximates gradients using central differences?
(A) Sobel
(B) Roberts
(C) LoG
(D) Canny
32. What is the main advantage of the Sobel operator over Prewitt?
(A) Simplicity
(B) Speed
(C) Better noise suppression
(D) Thin edges
33. Which edge detector is best suited for images with high noise levels?
(A) Canny
(B) Roberts
(C) Sobel
(D) Prewitt
34. Which of the following detects both positive and negative edges?
(A) Sobel
(B) Prewitt
(C) LoG
(D) Canny
35. Which operator has diagonal masks for gradient approximation?
(A) Roberts
(B) Sobel
(C) LoG
(D) Prewitt
36. In edge detection, the gradient is high when:
(A) Image is uniform
(B) Intensity changes rapidly
(C) Pixels are constant
(D) Image is blurry
37. What does hysteresis in Canny help prevent?
(A) Sharp edges
(B) Edge fragmentation
(C) Noise increase
(D) Smoothing loss
38. Which is not a commonly used edge detection method?
(A) Sobel
(B) Haar
(C) Canny
(D) Prewitt
39. Which of the following uses a single high-pass filter?
(A) Sobel
(B) LoG
(C) Canny
(D) Laplacian
40. Which edge detection method is considered multi-stage?
(A) Sobel
(B) Roberts
(C) Prewitt
(D) Canny
41. What kernel is used in Sobel for horizontal edge detection?
(A) [-1, 0, 1]
(B) [1, 1, 1]
(C) [0, -1, 0]
(D) [-1, -2, -1]
42. What is the role of Gaussian smoothing in Canny edge detection?
(A) Edge detection
(B) Gradient calculation
(C) Noise reduction
(D) Morphological transformation
43. Which edge detector has the best localization, detection, and low response?
(A) Sobel
(B) Prewitt
(C) Canny
(D) LoG
44. In Canny, edges are marked only if they are:
(A) Above high threshold
(B) Below low threshold
(C) Within the gradient
(D) Between two intensities
45. Which edge detection operator is based on zero-crossings?
(A) LoG
(B) Sobel
(C) Prewitt
(D) Roberts
46. Which operator emphasizes edges more heavily than Prewitt?
(A) Sobel
(B) LoG
(C) Canny
(D) Laplacian
47. Which edge detector requires tuning two thresholds?
(A) Sobel
(B) Prewitt
(C) Canny
(D) Roberts
48. Which operator computes gradient using orthogonal masks?
(A) Roberts
(B) Sobel
(C) Prewitt
(D) All of the above
49. What makes Canny different from other edge detectors?
(A) Simpler computation
(B) Use of color images
(C) Multi-step process with strong edge filtering
(D) No noise sensitivity
50. Which step is first in Canny edge detection?
(A) Double threshold
(B) Edge tracking
(C) Non-maximum suppression
(D) Gaussian smoothing
