VERK SUKENDRA SAI UNIVERSITY OF RU YUL 1 UYU yy erm Even/Odd Mid Semester Examination for session 2021-22 OURSE NAME: B-Tech. SEMESTER: 6” Electronics & Telecommunication Engineering Digital Image Processing ULL MARKS: 30 TIME: 90 Minutes ~~ Answersill Questions: ‘dicate Marks. Symbols carry usual meaning ‘The figures in the right hand margin in Ql. Answer all Questions. [2*3] a) Differentiate between image enhancement and image restoration. Give examples on ~ CO! each. b) Give the expression for logarithmic intensity transformation. For what kind of — = CO? ages such a transform is applicable for image enhancement? c) What is gray level slicing? Write its application. Draw transfer function of gray level = CO slicing on application of which the output image retains the background and other information except the region of interest. Q2. 18] 8) Consider the image segment as shown in Figure I. Compute the length of shortest 4-70! path, 8-path, and m-path between p & q assuming V = {1,2}. Also calculate the Ds. De distances between p and q. ; ~ ———— sy "Deine region in an image. Identify segions in the image given in Figure | considering 4-adjacency and V = {0,1}. oR a) Find 2x2 DCT transformation matrix and hence find the DCT of image -col mo Ire 2 (51 Poe One 2 b) Explain KL-transform. Write its properties and application. -co1 (3) Q. [8] a) Define H, $ and I in HSI color model, Draw HSI color model based on circular color -CO2 planes and triangular color planes. Indicate the planes at intensity 10.25, 0.5 and [5] 0.75 in both color planes (triangular and circular). Write the expression for converting from RGB color model to HS{ color model. b) Define_primary and. secondary colors Write the primary colors in light and -C0? a pigments. Illustrate how secondaryeol6rs are obtained in Tight and pigments using the primary colors. OR a) Define the term chromaticity. Draw C) ity diagram. Indicate th of red, green and blue in the chromatic ean - ae ae b) Explain CMY and CMYK color model in details[8] a) Given a32x32, 3-bit digital image with histogram data as given in the Table-1 -Co3 Perform histogram equalization and draw th: histogram of the equalized image. [5] b) A 4:bit image X shown if Figure 2 undergoes power law transformation with — - expression s=T(r); where, rand EP a Output grgy levels respectively. After transformation s is found 4o be $ %&.0911 “0.09g5 0.0179 ‘0.0367 0.0642 0.1012 0.1488 0.2077 0.2789 0.3629-0.4605 0.5724 0.6992 0.8416 $F Nter transformation the resultant output image is Y. Find ing 00 i 3 of the reg f image X : out the corresponding odtput pixel ‘Walyes of the region R of image X in th transformed image Y. oe en eS oR ibere. a) Determine the bit planes of a 6x4, 3-bit image given in Figure 3, Reconstruct the “ Wit 3- bit image using only two higher order bit planes. Sper e.een b) The histogram of a 3-bit image f(x,y) is shown in Figure 4. The lowest bit plane of — - C03 the image fis set to all zeros. Draw the new histogram of the resultant image. BI t 24) 6 Table 1 intensity] O [2 ]2 |3 [4 |5 |6 |7 8 ] | No.of | 220 | 110] 142] 56 | 59 | 90 | 125 | 224 Seana eee ee anne ---4 pixels: ar = | Figure 4 | Figure t 7 | [is op, i -| \Total Pages—6 Set-21(D B.Tech.-6th (ETC) Digital Image Processing Full Marks : 50 Time: 2. hours Answer all questions The figures in the right-hand margin indicate marks Symbols carry usual meaning Any supplementary materials to be provided 1. Answer all questions : 2 x5 (a) What do you mean by image interpolation ? (5) What is the principle objective of sharpe- ning spatial filters ? (c) Explain zero padding in an image. (d) Differentiate between constrained and unconstrained restoration. (e) Discuss about run length coding, ( Turn OnsC2) 2. ® Consider two-g bit images whose intensity levels span the full range from 0 to 255. Discuss the limiting effect of repeatedly subtracting image (2) from image (1). Assume that the result is represented in eight bits. : Bb Explain 4, 8 and m-adjacency along with connectivity of a binary image with suitable examples. 4 Or (a) Draw comparison between DFT and DCT algorithms in the context of image proce- ssing. State some of their properties and applications. 4 (6) Show that the 2-D transforms with separable, symmetric kernels can be computed by (1)-computing 1-D transforms along with individual rows (columns) of the input followed by computing 1-D transforms along the columns of the previous result. 4 B.Tech 6th (ETCYDigia! image p 2 (Continued )3. YY Explain power aw transformation « long with its significance by specifying suitable example. A Show that the second pass of the histogram equalization on the already histogram-equz- lized image will produce exactly the same result as the first pass. Or (a) Show. that if the coefficients of the spatial filter mask sum to zero, then the sum of all the resulting elements in the resulting convolution array will be also zero if you filter an image f(x, y) with a spatial filter mask w(x, y) using convolution. (6) Explain two techniques for image sharpening while showing proper examples for each of the techniques. Also discuss their advantages and disadvantages. 4. (@) Find out the DFT of the discrete function f(x,y) = cos(2mux + 2xvy),(4) (6) Consider a checkerboard image in which each square is 0.5 x 0.5 mm. Assuming that the image extends infinitely in both coordi- nate directions, what is the minimum sampling rate (in samples/mm) required to avoid aliasing ? 4 Or Bo List down the steps necessitated for filtering in frequency domain. Discuss some tech- “yen eniques-forimage sharpening using frequency domain filters. 4 (b) Consider a5 x § spatial mask that averages the 12 closest neighbors of a point (x, y) but excludes the point itself from the ave- rage. (i) Find the equivalent filter H (u,v) in the frequency domain. (di) Show that the resulting filter exhipits low pass charac- ! teristics. 4 5: MS With televant mathematical expressions, explain how a Wiener filter achieves resto- ration ofa Biven “ac ed image. B.Tech-6th (ETCYDigital Image a ( Const ) 4(5) es Explain techniques pertaining to restoration in presence of noise only-spatial filtering. 4 Or (a) What is the purpose of image restoration ? Explain the model of image degradation and restoration process using suitable block diagram. 4 (2) Discuss some of the techniques for estima- ting image degradation function. ~ 4 6. (a) What are the different image compression standards ? Explain. 4 (5) Describe arithmetic coding with an example for compression of image. Or 4 i i ¢ le i uffman coding with an examp goer of image. Explain Region Splitting and merging. stmage Processing(Se-21(D) (Twn Over ) evn Image 4(6) ‘(YK 1024 x 1024 8-bit image with 4.2 bits/ pixel entropy is to be Huffman coded. What ’ is the maximum compression that can be obtained ? 4