Tutorial 4 (UNIT–II)
Computer Vision and Image Processing
Branch–Section: CSE(AI&ML)–6CSM1
1. a) Define the concept of the Discrete Fourier Transform (DFT)?
The Discrete Fourier Transform (DFT) is a mathematical technique used in signal
processing and image analysis to convert a finite sequence of equally spaced samples of a
function into a sequence of complex numbers representing frequency components. The DFT
formula is given by:
X(k) = Σ [x(n) * e^(-j2πkn/N)] for n=0 to N-1
where X(k) represents the frequency domain, x(n) is the input sequence, N is the number of
samples, and j is the imaginary unit.
1. b) List the key properties of the 2-D Discrete Fourier Transform (2-D DFT) and
describe how they apply to image processing?
Key properties of 2-D DFT:
1. Linearity: The DFT of a sum of functions equals the sum of their DFTs.
2. Shift Theorem: A shift in the spatial domain corresponds to a phase shift in the frequency
domain.
3. Conjugate Symmetry: The DFT of real-valued images exhibits symmetry.
4. Periodicity: The DFT is periodic with a period equal to the size of the image.
5. Scaling: Scaling in one domain results in inverse scaling in the other domain.
These properties help in image compression, enhancement, and filtering operations.
2. a) Explain the relationship between the spatial domain and the frequency
domain in image processing, and how filtering in the frequency domain differs
from filtering in the spatial domain?
Answer: (Will be elaborated here with proper explanations, formulas, and diagrams if
necessary.)
2. b) Compare the Discrete Fourier Transform (DFT) and the Inverse Discrete
Fourier Transform (IDFT). How do they work together to convert signals
between the time domain and the frequency domain?
Answer: (Will be elaborated here with proper explanations, formulas, and diagrams if
necessary.)
3. a) Apply the Discrete Fourier Transform (DFT) to a set of sampled data and
demonstrate how the frequency components can be extracted. Show how
these components can then be used for noise reduction.
Answer: (Will be elaborated here with proper explanations, formulas, and diagrams if
necessary.)
3. b) Solve a problem where you need to filter an image in the frequency
domain using a low-pass filter. Utilize the DFT and IDFT to remove high-
frequency noise and then reconstruct the image in the spatial domain.
Answer: (Will be elaborated here with proper explanations, formulas, and diagrams if
necessary.)
4. a) Evaluate the effectiveness of frequency-domain filtering (e.g., low-pass,
high-pass filters) compared to spatial-domain filtering for noise reduction in
images. What are the advantages and limitations of each approach?
Answer: (Will be elaborated here with proper explanations, formulas, and diagrams if
necessary.)
4. b) Assess the impact of choosing different filtering techniques (e.g., Gaussian)
in the frequency domain for image smoothing. How do these choices influence
the quality of the final image and processing speed?
Answer: (Will be elaborated here with proper explanations, formulas, and diagrams if
necessary.)
5. a) Analyze the effect of applying a low-pass filter in the frequency domain to
an image. How does this transformation affect the smoothness and clarity of
the image, and what type of details (e.g., edges or noise) are lost?
Answer: (Will be elaborated here with proper explanations, formulas, and diagrams if
necessary.)
5. b) Compare the results of using a Gaussian filter versus a Butterworth filter
for image sharpening in the frequency domain. What are the key differences in
terms of the frequency components emphasized, and how do they impact the
final output?
Answer: (Will be elaborated here with proper explanations, formulas, and diagrams if
necessary.)
