The results obtained by a single pass through an image of some 2-D kernels can be achieved also by two passes using 1-D kernels. For example, the same result of using a 3 x 3 smoothing kernel with coefficients 1/9 can be obtained by a pass of the kernel [11 1] through an image, followed by a pass of the result with the kernel [11 1'. The final result is then scaled by 1/9. Show that the response of Sobel kernels (Fig. 10.14) can be implemented similarly by one pass of the differencing kernel [-1 0 1] (or its vertical coun- terpart) followed by the smoothing kernel [1 2 1] (or its vertical counterpart).

The results obtained by a single pass through an image of some 2-D kernels can be achieved also by two passes using 1-D kernels. For example, the same result of using a 3 x 3 smoothing kernel with coefficients 1/9 can be obtained by a pass of the kernel [11 1] through an image, followed by a pass of the result with the kernel [11 1'. The final result is then scaled by 1/9. Show that the response of Sobel kernels (Fig. 10.14) can be implemented similarly by one pass of the differencing kernel [-1 0 1] (or its vertical coun- terpart) followed by the smoothing kernel [1 2 1] (or its vertical counterpart).

Database System Concepts

7th Edition

ISBN:9780078022159

Author:Abraham Silberschatz Professor, Henry F. Korth, S. Sudarshan

Publisher:Abraham Silberschatz Professor, Henry F. Korth, S. Sudarshan

Chapter1: Introduction

Section: Chapter Questions

Problem 1PE

See similar textbooks

Similar questions

Consider a classifier trained on the 60k MNIST images (no augmentation). We test it on images where the pixels have been shifted. We expect the test accuracy to be poor. Is this an example of distribution shift? Please provide a reason.
Write the MATLAB code for the following
Compute the optimum threshold value k* for the given normalized histogram below using Otsu method. Note that image has four intensity levels only (i=0,1,2,3). Also, find the separability measure ƞ(k*).
Show that a high-pass filtered image can be obtained in the spatial domain as high pass = original-low pass, for simplicity assume 3 x 3 filters.
Please don't copy Describe a procedure for determining the pose of a robot in a 2-D world by using only “wall following” and assuming that a prefect map is available and that the absolute orientation of the robot can be determined at anytime.
a) We compute the gradient on an image using a differentiating kernel that we call Sobel kernel. We also prefer to run a Gaussian filter over the image before we compute the gradient. Explain (i) why we use a Gaussian filter before computing the gradient, (il) how we make this double filtering process computationally more efficient, and (ili) how we compute the gradient direction at a pixel. We threshold the gradient of the image twice when computing the Canny edges, using a low threshold and a high threshold. Explain why we use two thresholds instead of one, and what each threshold changes in the final result.
Please solve it , theoretical part of digital image processing course ; - If you have the following image, apply region growing from the shaded seed point using 8-connectivity and a predicate of “the absolute difference is less than or equal 25”.
Write Algorithm for Element from a Base ImageInput : a group G acting on f2 = { 1,2 ..... n };the base image [71, 72 ..... 7k[ of an element g of G;a base and strong generating set for G;Schreier vectors v (i) ,1 <_ i < k, for the stabiliser chain;Output : the element g = uk • Uk-1 X 9 9 9 X u 1;
Show working please
algorithm for backtrace proeeduregenerate( G : group; P : property; s : 1..k+l; l 9 1..k+l; i : 1..k+l;[71, 72 ..... 7i-1 ] : initial segment of base image;var K : group );(* Generate the coset representatives of G (t) in G (s) whose base image startwith [71, 72 ..... 7i-1] and .that may have property P.If one is found then extend K, the subgroup of G (s) of elementswith property P that have already been found, and return to search. *)