exam2soln - ECE/OPTI 533 Digital Image Processing Mid-Term...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
1/5 ECE/OPTI 533 Digital Image Processing Spring Semester, 2003 Mid-Term Exam 2 Solution May 6, 2003 Exam: Open-Notes, books, no partial credit Because there is no partial credit, it is especially important that you clearly defne, speciFy or describe everything that you do. 1. (10%,10%,10%) Given the following noisy image segment g(m,n): m = 1234 4 123 114 188 173 n = 3 120 126 193 199 2 135 127 197 201 1 182 179 189 193 The global image noise variance is 100, and the noise is additive to and independent of the noise- less image f(m,n). Calculate the adaptive Wiener Flter output at pixel (2,2) using a 3x3 pixel neighborhood and uniform box Flters. The adaptive W± output at pixel (m,n) is (notes17,p387): Using a 3x3 uniform box Flter, LP(2,2) = 160.89 HP(2,2) = 127 - 160.89 = -33.89 = 1075.86 (956.3 if variance calculated with N instead of N-1) therefore, = 975.86 and = 0.907 and = 130.15 (130.55) Note that, because this 3x3 region overlaps a high contrast feature and therefore has relatively high SNR, the change due to the W± is small. 2. (5%, 5%) Given the following 1-D image: f ˆ mn , () LP m n , σ f 2 , σ f 2 , σ v 2 + --------------------------------- HP m n , + = σ f 2 22 , σ v 2 + σ f 2 , σ f 2
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 5

exam2soln - ECE/OPTI 533 Digital Image Processing Mid-Term...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online