Scribd Logo
Upload

Welcome to Scribd!

  • 48 views

    In Spatial Domain: %% %low Pass Filter

    Uploaded by

    ripan030
    The document describes applying various spatial and frequency domain filters to an image, including low pass, Gaussian, weighted low pass, high pass, high boost, unsharp masking, Laplacian, and band pass filters. Code is provided to load an image, apply each type of filter using a convolution mask, and display the filtered output image for comparison to the original. Filters are applied both directly in the spatial domain and by filtering the image Fourier transform in the frequency domain.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    In Spatial Domain: %% %low Pass Filter

    Uploaded by

    ripan030
    48 views7 pages
    The document describes applying various spatial and frequency domain filters to an image, including low pass, Gaussian, weighted low pass, high pass, high boost, unsharp masking, Laplacian, and band pass filters. Code is provided to load an image, apply each type of filter using a convolution mask, and display the filtered output image for comparison to the original. Filters are applied both directly in the spatial domain and by filtering the image Fourier transform in the frequency domain.

    Original Description:

    different filter in image processing using mask

    Original Title

    ip3

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document describes applying various spatial and frequency domain filters to an image, including low pass, Gaussian, weighted low pass, high pass, high boost, unsharp masking, Laplacian, and band pass filters. Code is provided to load an image, apply each type of filter using a convolution mask, and display the filtered output image for comparison to the original. Filters are applied both directly in the spatial domain and by filtering the image Fourier transform in the frequency domain.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    48 views7 pages

    In Spatial Domain: %% %low Pass Filter

    Uploaded by

    ripan030
    The document describes applying various spatial and frequency domain filters to an image, including low pass, Gaussian, weighted low pass, high pass, high boost, unsharp masking, Laplacian, and band pass filters. Code is provided to load an image, apply each type of filter using a convolution mask, and display the filtered output image for comparison to the original. Filters are applied both directly in the spatial domain and by filtering the image Fourier transform in the frequency domain.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 7

    In spatial domain

    clc;clear all;close all;


    I=imread('C:\Users\Ripan\Desktop\sam.tif');
    imshow(I)
    title('Original image')
    %%
    %Low Pass Filter
    mask1=[1 1 1;1 1 1;1 1 1]/9;
    If1=imfilter(I,mask1);
    figure;
    imshow(If1)
    title('Low Pass Filtered image')
    %%
    %Gaussian Filter
    Ig=imnoise(I,'gaussian');
    figure;
    imshow(Ig)
    title('Image with Gaussian noise')
    mask2=zeros(3,3);
    p=1;
    for x=-1:1
    for y=-1:1
    mask2(x+2,y+2)=exp(-( x.^2+ y.^2 )/2*p^2);
    end
    end
    If2=imfilter(I,mask2);
    figure;
    imshow(If2)
    title('Gaussian Filtered image')
    %%
    %Weighted Low Pass Filter
    mask3=[1 2 1;2 4 2;1 2 1]/16;
    If3=imfilter(I,mask3);
    figure;
    imshow(If3)
    title('Weighted Low Pass Filtered image')
    %%
    %High Pass Filter
    mask4=[0 0 0;0 1 0;0 0 0]-[1 1 1;1 1 1;1 1 1]/9;
    If4=imfilter(I,mask4);
    figure;
    imshow(If4)
    title('High Pass Filtered image')
    %%
    %High Boost Filter
    mask5=3*[0 0 0;0 1 0;0 0 0]-[1 1 1;1 1 1;1 1 1]/9;
    If5=imfilter(I,mask5);
    figure;
    imshow(If5)
    title('High Boost Filtered image')
    %%
    %Unsharp Masking
    mask6=[0 0 0;0 1 0;0 0 0]+3*([0 0 0;0 1 0;0 0 0]-[1 1 1;1 1 1;1 1 1]/9);
    If6=imfilter(I,mask6);
    figure;
    imshow(If6)
    title('Unsharp Masking image')
    %%
    %Laplacian Filter
    mask7=[0 1 0;1 -4 1;0 1 0];

    If7=imfilter(I,mask7);
    figure;
    imshow(If7)
    title('Laplacian Filtered image')
    %%
    %Band Pass Filter
    mask8=[0 1 0;1 1 1;0 1 0]/5-[1 1 1;1 1 1;1 1 1]/9;
    If8=imfilter(I,mask8);
    figure;
    imshow(If8,[])
    title('Band Pass Filterd image')








    Original image
    Low Pass Filtered image
    Image with Gaussian noise Gaussian Filtered image




    Weighted Low Pass Filtered image
    High Pass Filtered image
    High Boost Filtered image
    Unsharp Masking image
    Laplacian Filtered image
    Band Pass Filterd image
    In frequency domain:

    clc;clear all;close all;
    I=imread('C:\Users\Girish\Desktop\sam.tif');
    figure;
    imshow(I)
    title('original image')
    %%
    %Low Pass Filtered image
    [r c]=size(I);
    rx=round(r/2);
    cx=round(c/2);
    If=fft2(I);
    H=zeros(r,c);
    h=[1 1 1;1 1 1;1 1 1]/9;
    for i=rx-1:rx+1
    for j=cx-1:cx+1
    H(i,j)=h(i-126,j-126);
    end
    end
    Hf=fft2(H);
    Imf=If.*Hf;
    Im=ifftshift(ifft2(Imf)/256);
    figure;
    imshow(Im)
    title('Low Pass Filtered image')
    %%
    %Gaussian Filter
    Ig=imnoise(I,'gaussian');
    figure;
    imshow(Ig)
    title('Image with Gaussian noise')
    Igf=fft2(Ig);
    h=zeros(3,3);
    p=1;
    for x=-1:1
    for y=-1:1
    h(x+2,y+2)=exp(-( x.^2+ y.^2 )/2*p^2);
    end
    end
    for i=rx-1:rx+1
    for j=cx-1:cx+1
    H(i,j)=h(i-126,j-126);
    end
    end
    Hf=fft2(H);
    Imf=Igf.*Hf;
    %Im=ifftshift(ifft2(Imf)/256);
    Im=ifftshift(ifft2(Imf)/256);
    figure;
    imshow(Im,[])
    title('Gaussian filtered image')
    %%
    %Weighted Low Pass Filtered image
    h=[1 2 1;2 4 2;1 2 1]/16;
    for i=rx-1:rx+1
    for j=cx-1:cx+1
    H(i,j)=h(i-126,j-126);
    end
    end
    Hf=fft2(H);
    Imf=If.*Hf;
    Im=ifftshift(ifft2(Imf)/256);
    figure;
    imshow(Im)
    title('Weighted Low Pass Filtered image')
    %%
    %High Pass Filtered image
    h=[0 0 0;0 1 0;0 0 0]-[1 1 1;1 1 1;1 1 1]/9;
    %h=h*9;
    for i=rx-1:rx+1
    for j=cx-1:cx+1
    H(i,j)=h(i-126,j-126);
    end
    end
    Hf=fft2(H);
    Imf=If.*Hf;
    Im=ifftshift(ifft2(Imf)/256);
    figure;
    imshow(Im)
    title('High Pass Filtered image')
    %%
    %High Boost Filtered image
    h=3*[0 0 0;0 1 0;0 0 0]-[1 1 1;1 1 1;1 1 1]/9;
    for i=rx-1:rx+1
    for j=cx-1:cx+1
    H(i,j)=h(i-126,j-126);
    end
    end
    Hf=fft2(H);
    Imf=If.*Hf;
    Im=ifftshift(ifft2(Imf)/256);
    figure;
    imshow(Im)
    title('High Boost Filtered image')
    %%
    %Unsharp masking
    h=[0 0 0;0 1 0;0 0 0]+3*([0 0 0;0 1 0;0 0 0]-[1 1 1;1 1 1;1 1 1]/9);
    for i=rx-1:rx+1
    for j=cx-1:cx+1
    H(i,j)=h(i-126,j-126);
    end
    end
    Hf=fft2(H);
    Imf=If.*Hf;
    Im=ifftshift(ifft2(Imf)/256);
    figure;
    imshow(Im)
    title('Unsharp masking image')
    %%
    %Laplacian Filtered image
    h=[0 1 0;1 -4 1;0 1 0];
    for i=rx-1:rx+1
    for j=cx-1:cx+1
    H(i,j)=h(i-126,j-126);
    end
    end
    Hf=fft2(H);
    Imf=If.*Hf;
    Im=ifftshift(ifft2(Imf)/256);
    figure;
    imshow(Im)
    title('Laplacian Fileredimage')
    %%
    %Band Pass Filtered image
    h=[0 1 0;1 1 1;0 1 0]/5-[1 1 1;1 1 1;1 1 1]/9;
    for i=rx-1:rx+1
    for j=cx-1:cx+1
    H(i,j)=h(i-126,j-126);
    end
    end
    Hf=fft2(H);
    Imf=If.*Hf;
    Im=ifftshift(ifft2(Imf)/64);
    figure;
    imshow(Im)
    title('Band Pass Filtered image')






    original image
    Low Pass Filtered image
    Image with Gaussian noise
    Gaussian filtered image



    Weighted Low Pass Filtered image
    High Pass Filtered image
    High Boost Filtered image
    Unsharp masking image
    Laplacian Fileredimage
    Band Pass Filtered image

    You might also like