Digital Image Processing — DU B.Sc. (Hons) Computer Science Previous Year Question Papers

Parts of a question must be answered together. 1 a) Define Isopreference curve with its application and suitable example. (2) b) Draw the bit planes for given 2-bit sub-image of size 2×2, (2) Sub-image c) Write any two sources of image noise. Write the mean and variance for (2) uniform noise. d) Calculate the number of bits required to store a 64 × 64 image with 128 (2) gray levels. (2) c) What is the significance of sampling? | low is the resolution of an image (2) affected by sampling? f) For the given 20-byte string, compute the encoded string by using run (4) length encoding. Also, calculate the compression ratio. SSSSSSSSSTTTTTNGGMMM g) Write 3x3 mask for the following low. pass filters used for image (4) i. Gaussian filter ii. Box filter h) Write all the steps involved in frequency domain filtering? (4) i). What is the significance of using a Laplacian filter on a digital image? (4) Apply Laplacian filter on the given image (I) on the center pixel using kernel (K) j) Answer the following: (4) i. Which morphological operation can be used to repair the broken characters in a digital image? ji. Write the name of the morphological operation which can be used to find the instances of object A in a digital image. ili. Write the name of the morphological: operation used to detect the shape of an object in a digital image. iv. Write the mathematical equation for the morphological operation used for boundary detection. a) Consider the following sub-image: (7) Let V = (0, 1) be the set of intensity values used to define adjacency. Compute the length of shortest of 4-, 8-, and m-path between p and q. If a particular path does not exist between these two points, explain why? b) Define Discrete Fourier Transform (DFT) and Inverse-Discrete Fourier 8) Transform (IDFT). Compute the DFT of the given time domain sequence: Values of exponentials are as follows: 3 a) Apply the Basic Global Thresholding algorithm on the following 4x4, (7) gray-scale sub-image to determine the final threshold value T and obtain the segmented binary sub-image. Sub-image Use the initial global threshold To=65. Stop the iteration when the threshold value is less than 1. Show all intermediate steps for each iteration. b) Consider the following 4-bit 5 x5 sub-image: (8) Perform the following operations for this sub-image: Negative of sub-image ii. Log transformation with c = / ili. Power law transformation with c = / and y = 5. 4 a) A source generates a sequence of symbols as follows: i. Determine the frequency and probability of each symbol. ii. Construct the Huffman code for the symbols based on their probabilities found in part (i). ili. Average Code Length of the source b) Given below is a 3 × 3 3-bit sub-image: Sub-image What would be the value of the center pixel when this image is passed through the following filters: i. Arithmetic mean filter Geometric mean filter iii. Harmonic mean filter iv. Min filter