Naotoshi Seo

Code: Spatial Filters

Table of Contents

Introduction Required tool Smooth Average Filter Median Filter Edge Detection Roberts Prewitt Sobel Laplacian LoG (Laplacian of Gaussian) Sharpen High Boost Filter

Introduction

Basic Image Processing tools. Spatial filters such as average filter, sobel filter, high boost filter, median filter.

Required tool

• 2-D Convolution cvpr:cvConv2.m
  • cvpr:cvuReflectBoundary.m
• 2-D Zero Crossing cvpr:cvZeroCross2.m

Smooth

Average Filter

cvpr:cvAverageFilter2.m

imshow(uint8(cvAverageFilter2(I, 3)));

Median Filter

Median filter is not a smoothing filter, but let me place at here because both median filter and smoothing filter is used for noisy images.

cvpr:cvMedianFilter2.m

imshow(uint8(cvMedianFilter2(I, 3)))

Edge Detection

Roberts

cvpr:cvRoberts.m

cvRoberts(I, 32, '45')

Prewitt

cvpr:cvPrewitt.m

Horizontal, Threshold 96

Vertical, Threshold 96

Sobel

cvpr:cvSobel.m

Horizontal, Threshold 96

Vertical, Threshold 96

Laplacian

cvpr:cvLaplacian.m

cvLaplacian(I, 200)

Threshold 200

LoG (Laplacian of Gaussian)

cvpr:cvLoG.m

Threshold 200

Sharpen

High Boost Filter

High boost filter is composed by an all pass filter and a edge detection filter (laplacian filter). Thus, it emphasizes edges and results in image sharpener.

cvpr:cvHighboostFilter2.m

Mask size m = 3, All pass factor A = 1, Division factor B = 1 (default)

Mask size m = 5, All pass factor A = 1, Division factor B = 1 (default)