Explore topic-wise InterviewSolutions in .

Right choice is (c) All of the mentioned

The BEST I can explain: SINCE, Laplacian is a derivative OPERATOR, so, highlights the gray-level DISCONTINUITIES in an image and deemphasizes areas with slowly varying gray levels. Hence, PRODUCES images having greyish edge lines superimposed on featureless background.

Correct option is (B) Mask that INCLUDES diagonal neighbors has more sharpness than the masks that doesn’t

For EXPLANATION I would say: Including diagonal NEIGHBOR pixels enhances sharpness of the image. So, Mask that includes diagonal neighbors has more sharpness than the masks that doesn’t.

Correct answer is (c) Because it contain both positive and negative VALUES

Explanation: A Laplacian filtered image contain both positive and negative values of COMPARABLE MAGNITUDES. So, SCALING is NECESSARY.

The correct option is (c) H(u, V)= [1 + (u – M/2)2+ (v – N/2)2].

To elaborate: The filter H(u, v)= [1 + (u – M/2)2+ (v – N/2)2] is used to perform the same enhancement in frequency DOMAIN LIKE in SPATIAL domain.

Right answer is (a) Fourier transform pair notation

Best EXPLANATION: The Fourier transform of the Laplacian result in spatial domain is EQUIVALENT to multiplying the F(u, v) and H(u, v). This dual relationship is expressed as Fourier transform pair notation GIVEN by: ∇^2 f(x,y)-[(u – M/2)^2+ (v – N/2)^2]F(u,v), for an IMAGE of size M *N.

Right CHOICE is (c) H(u, V)= -[(u – M/2)^2+ (v – N/2)^2].

Easiest explanation: The given OPERATION f(x, y)(-1)x+y shifts the center transform so that (u, v)=(0, 0) is at point (M/2, N/2) and HENCE the filter is: H(u, v)= -[(u – M/2)^2+ (v – N/2)^2].

The CORRECT choice is (c) ∇^2 f(x,y)↔-[(U –M/2)^2+ (V –N/2)^2]F(u,v)

Easiest explanation: The Fourier TRANSFORM of the Laplacian result in spatial domain is equivalent to multiplying the F(u, v) and H(u, v). This dual RELATIONSHIP is expressed as Fourier transform pair notation given by:∇^2 f(x,y)↔-[(u – M/2)^2+ (v – N/2)^2]F(u,v), for an image of size M*N.

Right choice is (B) (M/2, N/2)

The explanation: The given operation f(x, y)(-1)x+y SHIFTS the CENTER TRANSFORM so that (u, v)=(0, 0) is at point (M/2, N/2) for F and f of same size M*N.

Right choice is (c) Shifts the center transform

For explanation I WOULD say: The given operation f(x, y)(-1)x+y shifts the center transform so that (U, V)=(0,0) is at POINT (M/2, N/2) for F and f of same size M*N.

Right option is (c) H(u, v)= -(u^2+ v^2)

EASY EXPLANATION: LAPLACIAN in FREQUENCY domain is: I[(∂^2 f(x,y))/∂x^2 +(∂^2 f(x,y))/∂y^2 ]= -(u^2+v^2)F(u,v), where ℑ is the Fourier transform operator and F(u, v) is Fourier transformed function of f(x, y) and -(u^2+ v^2) is the filter.

Right answer is (B) Filtering operation

To explain I would say: The Laplacian in FREQUENCY DOMAIN is simply implemented by USING filter:

H(u, V)= -(u^2+ v^2).

Right OPTION is (c) All of the mentioned

Easy explanation: Since gradient are used in fist order derivative IMAGE enhancement that enhances the DISCONTINUITIES except for in flat areas and PRODUCES thick edge for CONSTANT slope ramp. So, Gradient has all the mentioned features.

The correct option is (a) MAGNITUDE of Gradient vector

Best explanation: Magnitude of Gradient vector is USED for implementation of FIRST derivative in image processing, while LAPLACIAN is for second order implementation in image processing.

Right option is (c) 0

The best explanation: Since, first ORDER DERIVATIVE of a DIGITAL function must be zero in the areas of constant grey values. So, the mask using gradient has a sum 0, so to produce a zero result if APPLIED on constant gray level areas.

The correct choice is (a) True

The best explanation: As the value of A increases, sharpening process contribution becomes LESS IMPORTANT and so at some very LARGE value A, the contribution becomes ALMOST negligible and so high boost filtered image is APPROXIMATELY equal to the original image.

The CORRECT OPTION is (b) 1

The best explanation: for A=1 the HIGH BOOST filtering is given by:

.

Correct ANSWER is (c) All of the mentioned

For explanation: Unsharp MASK sharpens IMAGES by subtracting a BLURRED version of original IMAGE from the original image itself.

A high-boost filter is a generalized form of unsharp mask.

The correct ANSWER is (a) FHB = A f(X, y) – ∇^2f(x,y)

The explanation: If LAPLACIAN is used to obtain sharp image for unsharp MASK filtered image, then

.

The correct ANSWER is (d) All of the mentioned

The explanation is: A high-boost FILTER is a generalized FORM of unsharp mask and is GIVEN by:

 fhb = A f(x, y) – f (x, y)

 Or, fhb = (A – 1) f(x, y) + f(x, y) – f(x, y), that can be written as

 fhb = (A – 1) f(x, y) + FS(x, y), where fs(x, y) = f(x, y) – f (x, y).

The correct choice is (a) 5

For explanation: The mask FORMED by eliminating DIAGONAL neighbors i.e. 4f(x, y), SINCE each diagonal CONTAIN a -2f(x, y), the mask has 5 as its CENTRAL coefficient.

Right choice is (C) Linear operator

Best explanation: DERIVATIVE of any order are linear OPERATIONS and SINCE, Laplacian is the simplest ISOTROPIC derivative operator, so is a linear operator.

Order-Statistics operator are nonlinear operators.

Correct option is (a) If DEFINITION used has a negative center COEFFICIENT, then subtraction is done

To explain: Applying Laplacian PRODUCES image having featureless background which is recovered maintaining the SHARPNESS of Laplacian operation using original image either added if Laplacian definition used has a positive center coefficientor subtracting RESULT from original image if has a negative center coefficient.

The CORRECT option is (a) 90^o

The explanation is: The GIVEN Laplacian GIVES ISOTROPIC result for 90^o INCREMENTAL rotations.

The correct answer is (a) [f(x + 1, y) + f(x – 1, y) – 2F(x, y)] and [f(x, y + 1) + f(x, y – 1) – 2f(x, y)] respectively

The explanation is: For a LAPLACIAN given by:∇^2 f=

Applying second order DERIVATIVE in x DIRECTION (∂^2 f)/∂x^2= [f(x + 1, y) + f(x – 1, y) – 2f(x, y)], and

Applying second order derivative in y direction (∂^2 f)/∂y^2 = [f(x, y + 1) + f(x, y – 1) – 2f(x, y)].

Right option is (a) Laplacian

The BEST explanation: An ISOTROPIC filtering is an example of second order derivative for ENHANCEMENT and uses Laplacian as the simplest derivative operator, while gradient is used with FIRST derivatives.

The correct option is (a) Increase in K, decreases the noise of pixel values

To explain I would say: As K increases, E {ḡ(x, y)} the expected VALUE approaches f(x, y) the ORIGINAL IMAGE, i.e. decreasing the noise COMPONENT.

Correct CHOICE is (a) Isotropic filters

Explanation: Isotropic FILTER are ROTATION invariant because it has a same response when APPLIED to the IMAGE first and the after rotating the image.

The correct answer is (d) NONE of the mentioned

Easy EXPLANATION: In Image Averaging enhancement METHOD assumptions are made for a noisy image g(x, y) that at every coordinate (x, y) the noise has 0 AVERAGE value and MUST be uncorrelated.

The CORRECT choice is (a) Their covariance is 0

The best I can explain: The covariance of two random VARIABLES x i and x J given by: E [(x i – m i) (x j – mj)], E {.} is EXPECTED value of the ARGUMENT and m is the mean. If this covariance turns out to 0, the variables are uncorrelated.

The correct option is (C) The truncation inherent in division by 2 causes loss in ACCURACY

Explanation: The GIVEN method is quite simple and easy to IMPLEMENT, however it has the lim itation of accuracy loss because of truncation inherent in division by 2 and also that it doesn’t ENSURE the full range usage.

The correct option is (d) All of the mentioned

Easy explanation: Proper Registration does special marking into the PRODUCT in case two IMAGES are identical so as the difference won’t create any sense.

Controlled Illum ination is important because changes in illum ination can affect dramatically the difference image values.

Noise levels of a difference image must low enough so that the variation due to noise won’t affect the difference VALUE much.

The correct option is (d) None of the mentioned

Easy EXPLANATION: The RANGE of a result of a subtracted image is -255 m inimum to 255 max imum, if 8-bit channel is USED to display the original images.

Correct option is (a) Because the difference in such AREAS is little, that yields low value

The EXPLANATION is: There remains a little CHANGE in some areas in the images to be SUBTRACTED that yields low value and so the result appears as dark shades of gray.

Correct option is (b) SUBTRACTION

Easy explanation: In the given area of medical imaging, a mask of an X-ray image of a region of subject is captured using TV camera is subtracted from image of same region taken after injecting a contrast medium to the BLOODSTREAM. The subtraction result GIVES an ENHANCED detail of how a contrast medium propagates through the bloodstream.

This the best commercially successful METHOD.

Correct option is (c) 00000100

For EXPLANATION: For AND operation results in 1 only for 1AND 1, else 0.

All the BITS of the GIVEN GRAY value are operated similar resulting in 00000100.

The CORRECT option is (c) MULTIPLICATION

To ELABORATE: Multiplication of one image by ANOTHER is USED as a gray-level mask.

Right CHOICE is (c) NOT, OR

To elaborate: AND, OR operators are used for MASKING, while NOT works as NEGATIVE transformation.

The correct option is (c) NOT

Best explanation: APPLYING NOT operator on a BLACK, 8-bit pixel gives a white, 8-bit pixel, so, is equivalent to NEGATIVE TRANSFORMATION.

Right OPTION is (a) A string: 00000000

The best explanation: LOGIC operation on gray-scale images are done by processing of pixel VALUES as string of binary NUMBERS, so, a black, 8-bit pixel is processed as a string of eight 0’s.

Right option is (C) String of binary numbers

Explanation: Logic OPERATION on gray-scale images are done by processing of pixel VALUES as string of binary numbers.

The correct choice is (d) All of the mentioned

Best explanation: All the THREE LOGICAL operators given are functionally COMPLETE because all other logical operators can be IMPLEMENTED using these three.