EXAM SOLUTIONS Image Processing and （图像处理和考试解决方案）.pdfVIP

School of Computer Science and Communication, KTH Lecturer: Danica Kragic EXAM SOLUTIONS Image Processing and Computer Vision, DD2423 Tuesday, 16th of March 2010, 14.00–19.00 Grading: 45 - 50 A 34 - 44 B 23 - 33 C 12 - 22 D 0 - 11 E 1 Part A Provide short answers to the questions! Each answer is worth maximum one point. 1. Explain briefly the difference between convolution and correlation. These operations have two key features: they are shift-invariant and linear. Differences should have been shown either by prvoviding the formula fro each or stating that the process is really the same except that we flip over the filter before correlating. 2. What are the two image processing processes performed by the Laplacian of Gaussian operator. smoothing and 2nd order differentiation 3. Draw two histograms: one of a high contrast image and one of a low contrast image. High contrast histogram commonly covers the whole spectra of the intenisties, while low contrast histogram is very narrow. 4. Give an example of an isotropic filter. Any rotationally invariant filter. 5. What is disparity in stereo? The difference in vertical or horizontal position of a perceived 3D point between two images. 6. Write a Sobel filter mask and explain what it consists of. You needed to provide a mask and show that it consists of weighted averaging and a derivative filter. 7. Explain the terms sampling and quantization. Sampling: discretization of a signal (spatial or temporal) Quantization: assigning (in image processing) an intensity value to the signal 8. What is “contrast reversal” in terms of gray-level transformations? Draw the corresponding linear transforma-