Scribd
Upload
11 views

Lab 5

Here is the code to implement an average filter on an image: % Read color image I = imread('image.jpg'); % Convert to grayscale Ig = rgb2gray(I); % Generate average filter mask h = ones(5)/25; % Apply filter to grayscale image Iavg = imfilter(Ig,h,'replicate'); % Add salt and pepper noise Inoise = imnoise(Ig,'salt & pepper',0.5); % Apply filter to noisy image Iavg_noise = imfilter(Inoise,h,'replicate'); % Display results figure; subplot(2,2,1

Uploaded by

Supriya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
11 views

Lab 5

Here is the code to implement an average filter on an image: % Read color image I = imread('image.jpg'); % Convert to grayscale Ig = rgb2gray(I); % Generate average filter mask h = ones(5)/25; % Apply filter to grayscale image Iavg = imfilter(Ig,h,'replicate'); % Add salt and pepper noise Inoise = imnoise(Ig,'salt & pepper',0.5); % Apply filter to noisy image Iavg_noise = imfilter(Inoise,h,'replicate'); % Display results figure; subplot(2,2,1

Uploaded by

Supriya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 11

SPATIAL FILTER

Average Filter Implementation


Median Filter Implementation
Gaussian Filter Implementation

1
SPATIAL FILTER- Filters

Commands used : imread( ),imshow(), subplot/figure


rgb2gray, vector generation, sum function ,ones,

2
imread() & imshow()

 Reads a jpg file stores in drive or file.


A1= imread(‘try1.jpg’)

imshow(image1)
Displays the grayscale image

imshow(image1,[])
Displays the grayscale image with max and min range
subplot() & figure()

subplot(m,n,x)
m ---No. of rows
n --- No. of columns
x --- Figure no.
subplot(3,1,2)
m=3, n= 1 x=2
figure( 2)
2: handler
Vector generation
 a= [ 1 2 3 4 ];
A1= [ 1 2 3 4 ;
2 4 6 8;
3 6 9 12;
4 8 12 16 ]
 average generation : n= [ 1 2 3 4]
sum (n) ( sums all functions row wise)
N=sum( A1)
B= ones(4)
B1= ones(4,4) %B=B1
conv2
• avgfiltmask= A1/N
• Data: double
• Convert to double
• double(avgfiltmask) double(image2)
• Convolution output is single if one input or both is single.Else
double
• conv : image ( M*N) full,same,valid
• 'full' — function gives the full 2-D convolution.( M+N-1)
• 'same' —central part of the convolution, which is the same size
as A.(M*N)
• 'valid' — convolution without zero-padded edges.
• For a filter of size (2K+1) × (2L+1), if the image size is MxN, the
filter is computed over the range:
• K=1(3×3), L=1 (3×3) M×N = 256×256
• 1≤u’ ≤ 254 (256-1-1) 1≤v’ ≤ 254 (256-1-1)
Task
 Read an image, check its size, convert to gray scale and set as
double and as the same image. Plot the image.
 Build an average filter of size( 5*5) using ones function and
using vector generation.Plot the filter.
 Convolve the filter and observe output.Plot all outputs. rgb
image,Grayscale img, convolved image 1, convolved image 2
 Plot 4 images.Verify that the last and before last outputs
( both result of convolved images) are equal.
Salt and Pepper Noise
• Form of impulse noise( sudden disturbances, sensor faults)
• Eg. MRI scan
• Certain amount of the pixels in the image are either black
or white
• Random bright (with 255 pixel value) & Random dark
(with 0 pixel value)
• Data drop noise in statistics
• Drop the original data values
Salt and Pepper Noise

• imnoise function
• A = imnoise(image1,'salt & pepper‘,val);
• val: density of noise, default value is 5% of total
pixels present in I

9
Average Filter Implementation Algorithm

 Read the colour image using imread( ).Resize image


Convert to gray using rgb2gray
 Generate the average filter mask
 Apply convolution to the gray image
Observe output using the function imshow( )
Add salt and pepper noise( 0.5) to the gray image
Apply convolution to the noisy gray image
Observe output using the function imshow( )
Average Filter Code

You might also like