Explore topic-wise InterviewSolutions in .

Right CHOICE is (c) All of the mentioned combined and applied

The BEST explanation: To accentuate the CONTRIBUTION to enhancement made by high-frequency components, we have to multiply the HIGHPASS FILTER by a constant and add an offset to the highpass filter to prevent eliminating zero frequency term by filter.

Right OPTION is (a) MASK that excludes the diagonal NEIGHBORS

Best explanation: The frequency DOMAIN LAPLACIAN is closer to mask that excludes the diagonal neighbors.

Correct choice is (a) Composite filter

To explain I would SAY: UNSHARP masking can be IMPLEMENTED directly in frequency DOMAIN by using a composite filter: Hhp(u, v) = 1 – Hlp(u, v).

Correct choice is (d) Hhp(u, v) = 1 – Hlp(u, v)

For explanation: Unsharp masking can be implemented DIRECTLY in frequency DOMAIN by USING a composite FILTER: Hhp(u, v) = 1 – Hlp(u, v).

Right CHOICE is (a) HIGH-boost filtering reduces to regular Highpass filtering

The EXPLANATION: High boost filtered IMAGE is MODIFIED as: fhb = (A-1) f(x, y) +f(x, y) – flp(x, y)

i.e.fhb = (A-1) f(x, y) + fhp(x, y). So, when A=1, High-boost filtering reduces to regular Highpass filtering.

Right answer is (b) High-boost filtering

To ELABORATE: UNSHARP masking is defined as “obtaining a highpass filtered image by SUBTRACTING from the given image a lowpass filtered version of itself” while high-boost filtering GENERALIZES it by multiplying the input image by a constant, say A≥1.

The correct option is (c) All of the mentioned

The EXPLANATION: High BOOST filtered image is modified as: fhb = (A-1) F(x, y) +f(x, y) – FLP(x, y)

i.e.fhb = (A-1) f(x, y) + fhp(x, y). So, when A>1, the contribution of the image itself becomes more dominant over the highpass filtered version of image.

The correct choice is (a) The narrower the frequency DOMAIN filter more severe is the ringing

To ELABORATE: The characteristics feature of reciprocal relationship says that the narrower the frequency domain filter becomes it attenuates more low frequency component and so INCREASES blurring and more severe becomes the ringing.

Right choice is (d) None of the mentioned

The EXPLANATION is: The center COMPONENT of ideal LOWPASS filter is primarily RESPONSIBLE for blurring while, concentric component is primarily for the ringing CHARACTERISTIC of ideal filter.

The CORRECT option is (c) All of the mentioned

Easy explanation: the lowpass FILTER has two different characteristics: one is a dominant COMPONENT at origin and other one is a CONCENTRIC, circular components about the center component.

The CORRECT choice is (b) Gaussian lowpass filter

Explanation: For high VALUE of filter ORDER Butterworth lowpass filter behaves as Ideal lowpass filter, while for lower order value it has a smoother FORM behaving LIKE Gaussian lowpass filter.

Right answer is (b) FILTER size is inversely proportional to BLURRING caused because of filter

Easiest explanation: Increase in filter size, removes LESS power from the image and so less SEVERE blurring occurs.

Correct ANSWER is (a) Ideal filter

For explanation I WOULD say: In ideal filter all the frequencies inside a CIRCLE of radius D0 are not attenuated while all frequencies OUTSIDE the circle are completely attenuated.

Correct option is (a) Ideal lowpass FILTER

The BEST explanation: For high value of filter ORDER Butterworth lowpass filter behaves as Ideal lowpass filter, while for LOWER order value it has a smoother form behaving like GAUSSIAN lowpass filter.

Right choice is (a) IDEAL lowpass filters

To explain I WOULD say: Ideal lowpass FILTER COVERS the range of very SHARP filter functioning of lowpass filters.

Right option is (d) All of the mentioned

Explanation: Lowpass FILTERS are considered of THREE TYPES: IDEAL, Butterworth, and GAUSSIAN.

Correct answer is (a) The NARROWER the FREQUENCY DOMAIN filter results in increased blurring

Explanation: The characteristics feature of reciprocal relationship says that the narrower the frequency domain filter becomes it ATTENUATES more low frequency component and so INCREASES blurring.

The correct OPTION is (a) Fourier TRANSFORM

Easiest EXPLANATION: Filters in spatial domain and frequency domain has a Fourier transform PAIR relation.A frequency domain filter of the corresponding filter in spatial domain can be OBTAINED by applying inverse Fourier transform on spatial domain filter.

Right option is (B) Inverse Fourier transform

The best I can explain: FILTERS in spatial domain and FREQUENCY domain has a Fourier transform pair relation.A spatial domain filter of the CORRESPONDING filter in frequency domain can be OBTAINED by applying inverse Fourier transform on frequency domain filter.

Right choice is (d) All of the mentioned

Explanation: A HIGHPASS filter attenuates low frequency so have less gray-level variation in smooth areas, and ALLOWS high frequencies so have EMPHASIZED TRANSITIONAL gray-level details, resulting in a SHARPER image.

Correct answer is (b) LOWPASS filter

The best explanation: A lowpass filter attenuates high so the image has less SHARP DETAILS.

Right option is (a) F(0, 0)

BEST explanation: For an image f(x, y), the FOURIER transform at origin of an image, F(0, 0), is equal to the average VALUE of the image.

The correct answer is (b) NOTCH FILTER

Best EXPLANATION: Notch filter sets F (0, 0), to zero, hence setting up the average value of image zero. The filter is named so, because it is a constant function with a notch at origin and so is able to SET F (0, 0) to zero leaving out other values.

Correct answer is (C) The resultant image will be a little smaller SIZE than ORIGINAL image

Explanation: If the center of mask MUST be at a DISTANCE ≥ (n – 1)/2 pixels from border of image, the border pixels won’t get processed under mask and so the resultant image would be of smaller size.

The correct choice is (c) Zero-phase-shift-filter

The explanation is: Zero-phase-shift-filter multiplies two functions F(u, v) and H(u, v), where F has complex components too SINCE is FOURIER TRANSFORMED function of f(x, y), in an order that each component of H multiplies both REAL and complex part of corresponding component in F.

The correct CHOICE is (B) Computation of median

To explain I would say: Computation of median gray-level value in the neighborhood is the basic function of NONLINEAR FILTER for noise REDUCTION.

The correct choice is (b) Moving away from the origin of TRANSFORM the higher FREQUENCIES corresponds to abrupt CHANGE in gray level

Explanation: Moving away from the origin of transform the LOW FREQUENCY corresponds to the slowly varying components in an image. Moving further away from origin the higher frequencies corresponds to faster gray level changes.

Right choice is (C) All of the mentioned

To explain: Moving AWAY from the origin of transform the low frequency corresponds to the slowly varying components in an image. Moving further away from origin the higher frequencies corresponds to faster GRAY level changes.

The CORRECT option is (c) All of the mentioned

Explanation: COMPUTATION of VARIANCE as well as median comes under nonlinear OPERATION.

Right choice is (a) Sum of filter coefficient’s product and corresponding image pixel under filter mask

Easiest explanation: In SPATIAL filtering the mask is moved from POINT to point and at each point the response is CALCULATED using a PREDEFINED relationship. The relationship in linear spatial filtering is GIVEN by: the Sum of filter coefficient’s product and corresponding image pixel in area under filter mask.

Correct answer is (d) All of the mentioned

Explanation: Working in neighborhood operations is DONE with the VALUE of a subimage having same dimension as neighborhood corresponding to the value in the image PIXEL. The subimage is called as filter, MASK, TEMPLATE, kernel or window.

The CORRECT choice is (c) 50th percentile

The explanation: Since the median filter forces median intensity to the PIXEL which is ALMOST the LARGEST value in the middle of the list of values as PER the ranking, so represents a 50th percentile ranked set of numbers.

Right choice is (C) All of the mentioned

The EXPLANATION: The isolated cluster pixel value doesn’t come as a median value and since are EITHER are light or dark as compared to neighbors, so are forced with median intensity of neighbors that aren’t EVEN close to their original value and so are sometimes termed “eliminated”.

If the area of such isolated PIXELS are < n2/2, that is again the pixel value won’t be a median value and so are eliminated.

Larger cluster pixels value are more pronounced to be a median value, so are considerably less forced to median intensity.

Right answer is (b) Max filter

The EXPLANATION is: A max filter gives the BRIGHTEST point in an image and so is USED.

Correct option is (b) 20

For explanation I would say: The values are FIRST sorted and so turns out to (10, 15, 20, 20, 20, 20, 20, 25, and 100). For a 3*3 NEIGHBORHOOD the 5th LARGEST value is the MEDIAN, and so is 20.

Right option is (C) All of the mentioned

The explanation: An impulse noise has an appearance as black and white dots superimposed on the IMAGE. This is also KNOWN as Salt-and-pepper noise.

Right option is (d) All of the mentioned

Explanation: An Order-Statistic FILTERS also CALLED non-linear spatial filters, RESPONSE is based on ranking the pixel in the image encompassed by filter that replaces the CENTRAL pixel value. A Median filter is an example of such filters.

Correct choice is (a) Median filter EFFECTS in considerably less blurring than the LINEAR spatial filters

For explanation: For IMPULSE NOISE, median filter is much effective for noise reduction and causes considerably less blurring than the linear spatial filters.

The CORRECT answer is (a) True

The best EXPLANATION: A MEDIAN FILTER the pixel value is replaced by median of the gray-level in the neighborhood of that pixel and also the original pixel value is included while computing the median.

Correct answer is (d) Blurring increases with increase of the size of FILTER size

Explanation: USING a size 3 filter 3*3 and 5*5 size squares and other OBJECTS shows a significant blurring with respect to object of larger size.

The blurring gets more pronounced while using filter size 5, 9 and so on.

Right option is (a) It results in increasing blurring in SMOOTHING process

Easy EXPLANATION: Use of a mask having central coefficient maximum and then the coefficients reducing as a function of increasing distance from ORIGIN is a STRATEGY to REDUCE blurring in smoothing process.