Image Degradation and Restoration Quiz

Which of the following best describes the image degradation model?

• g(x,y) = f(x,y) - h(x,y) + η(x,y)

• g(x,y) = h(x,y) * f(x,y) + η(x,y)

• g(x,y) = f(x,y) / h(x,y)

• g(x,y) = h(x,y) + η(x,y)

In the degradation model, what does η(x,y) represent?

• Original image

• Blurring function

• Additive noise

• Filter kernel

What is assumed in a linear position-invariant degradation model?

• Blur changes with image position

• Noise is multiplicative

• Degradation process does not vary spatially

• Noise depends on time

In frequency domain, degradation can be represented as:

• G(u,v) = F(u,v) / H(u,v)

• G(u,v) = F(u,v) * H(u,v) + N(u,v)

• G(u,v) = F(u,v) + H(u,v)

• G(u,v) = H(u,v) - F(u,v)

Which filter directly tries to reverse the degradation function in frequency domain?

• Wiener filter

• Constrained least squares

• Inverse filter

• Laplacian filter

What is a major limitation of inverse filtering?

• Works only in spatial domain

• Does not remove blur

• Amplifies noise

• Cannot process grayscale images

Wiener filtering is based on which criterion?

• Maximum entropy

• Minimum absolute error

• Minimum mean square error

• Maximum likelihood

Wiener filter requires knowledge of:

• Only degradation function

• Noise power spectrum and image power spectrum

• Only original image

• Only blurred image

Which domain is most commonly used for Wiener filtering?

• Spatial domain

• Gradient domain

• Frequency domain

• Time domain

Constrained least squares filtering differs from Wiener filtering by:

• Ignoring noise

• Including a constraint on smoothness

• Working only in time domain

• Using median filters