spatial filter

24
SPATIAL DOMAIN FILTERING 1

Upload: shalet-k-s

Post on 19-Jul-2015

137 views

Category:

Documents


7 download

TRANSCRIPT

SPATIAL DOMAIN

FILTERING

1

2

What are images?

• 2D image f(x,y).

•x and y are spatial co-ordinates.

•Amplitude of f at any (x,y) is called intensity or

gray level of the image.

3

•Continuous image to discrete image

4

•Result is a matrix of real numbers of size M*N.

•Each element in this array is called a pixel.

5

OBJECTIVE

The principal objective of enhancement is to process

an images so that the result is more suitable than the

original image for a SPECIFIC application.

Category of image enhancement

• Spatial domain

• Frequency domain

6

WHAT IS???

Spatial filtering (neighborhood processing) is defined by:

A neighborhood

An operation that is performed on the pixels inside

the neighborhood

7

•Typically, the neighborhood is rectangular and its

size is much smaller than that of f(x , y)

e.g., 3x3 or 5x5

Neighborhood

8

Neighborhood

•Coefficients of neighborhood is also called as filter,

mask, filter mask, kernel, template or window.

•Image size is M*N and filter mask size m*n.

•m=2a+1 and n=2b+1.

•a and b are non-negative integers.

9

Assume the

origin of the

mask is the

center of the

mask.

for a 3 x 3 mask,

1 1

1 1

( , ) ( , ) ( , )s t

g x y w s t f x s y t

10

LINEAR V/S NON-LINEAR

A filtering method is linear when the output is

a weighted sum of the input pixels.

Methods that do not satisfy the above property

are called non-linear.

11

Linear Spatial Filtering Methods

Two main linear spatial filtering methods:

Correlation

Convolution

12

Correlation:

•Process of passing the mask w by the image array f.

Convolution

•Often used in applications where we need to

measure the similarity between images or parts of

images

(e.g., pattern matching).

•Similar to correlation except that the mask is first

flipped both horizontally and vertically.

Note: if w(x,y) is symmetric, that is w(x,y)=w(-x,-y),

then convolution is equivalent to correlation!

13

TOOL BOX IMPLEMENTATION

g = imfilter(f, w, filtering_mode, boundary_options,

size_options)

Common syntax

g = imfilter(f, w, ‘replicate’)

14

How do we choose the elements of a

mask?

Typically, by sampling certain functions.

Gaussian1st derivativeof Gaussian

2nd derivativeof Gaussian

Gaussian1st derivativeof Gaussian

2nd derivativeof Gaussian

15

Spatial filters : Smoothing ( low pass)

Use: for blurring and noise reduction.

How it works? The value of every pixel is replaced by the

average of the gray levels in the neighborhood.

Type of smoothing filters:

1. Standard average

2. Weighted average.

3. Median filter

Gaussian

16

1 1 1

1 1 1

1 1 1

9

11 2 1

2 4 2

1 2 1

16

1

Standard average Weighted average13090120110

200989491

100999190

90859682

Standard averaging filter:

(110 +120+90+91+94+98+90+91+99)/9 =883/9 = 98.1

Weighted averaging filter:

(110 +2 x 120+90+2 x 91+4 x 94+2 x 98+90+2 x 91+99)/16 =97.81

17

Mask size determines the degree of smoothing and

loss of detail.

original 3x3 5x5 7x7

15x15 25x25

18

13090120110

200989491

100999590

90859682

95

Steps:

1. Sort the pixels in ascending order:

90,90, 91, 94, 95, 98, 99, 110, 120

2. replace the original pixel value by the median : 95

Median filter(non linear)

19

Very effective for removing “salt and pepper” noise

(i.e., random occurrences of black and white pixels).

Averaging Median filtering

20

Spatial filters : Sharpening ( high pass)

Use: for highlighting fine detail or enhancing

detail that has been blurred.

1st derivativeof Gaussian

1st derivativeof Gaussian

2nd derivativeof Gaussian

Type of sharpening filters:

21

155153156157153

159156158156159

160156154158155

160160158157154

155156157157157

•154*4 – 158- 156-158-158 = -14

•So the value after filter = -14

•We call the resultant image: sharpened image.

•Filtered image=original +sharpened image

•The value in the filter image=154-14 =130

SHARPENING – 1ST DERIVATIVE

•Apply the following laplace on the highlighted pixel

22

155153156157153

159156158156159

160156154158155

160160158157154

155156157157157

•Apply the following laplace 2nd derivative on the highlighted pixel

•154*4 – 158- 156-158-158 = -14

•Apply laplace to all pixels

•Then apply it again to our pixel:-14*4 – 10 -10 – (-6) -4 =-74

•So the value after 2nd derivative filter =-74

•The value of pixel in the filter image=154-74 = 80

SHARPENING – 2nd DERIVATIVE

23

Original Image

Laplace Sharpened Image

Laplace Filtered Image

24

Thank you