You are on page 1of 1

Computer Vision: Assignment-3

Topic: Image Restoration

1. What are the factors that can cause image degradation? Explain in detail an image
degradation and restoration model.
2. What is the role of the point spread function? List out some of the kernels or masks of the
PSF.
3. What is noise? How can noises be classified? Explain the PDFs of various noise patterns and
define their noise estimation parameters (mean and variance).
4. Write about the types of mean filters and noises they remove from images.
5. What is an order-statistic filter? How is it different from convolution-based filters? What is
the result of applying an order-statistic filter on the following image?
1 3 5
[4 4 3]
5 2 2
6. Consider the following list of elements: A = {15, 15, 25, 25, 79, 80, 80,110}
Apply alpha-trimming filter with a clipping factor T=2 find the replacing pixel value.
7. What is Wiener filter? Write in detail the construction and implementation of a Wiener filter.
List out the advantages and disadvantages of a Wiener filter.
8. Discuss the homomorphic filtering approach for removing speckle or multiplicative noises
from the digital images.
9. Discuss the concepts of high boost filtering, unsharp masking and crispening and their
applications.
10. Discuss the mathematical concepts of following:
A) Inverse filtering
B) Constrained least square filters
C) Tikhonov Regularization
11. Define the following image quality metrics used to measure the performance of restoration
and enhancement filters:
A) Mean Square Error (MSE) and Root Mean Square Error (RMSE)
B) Absolute error
C) Maximum absolute deviation (MAD)
D) Contrast
E) Signal-to-Noise Ratio (SNR)
F) Peak Signal-to-Noise Ratio (PSNR)
G) Correlation parameter
H) Mean Structure Similarity Index Map (MSSIM)

You might also like