IMAGE PROCESSING

Sobel edge detection

The gradient of the image is calculated for each pixel position in the image.

The procedure and the MATLAB code for sobel edge detection without using MATLAB built-in function:

MATLAB CODE:

A=imread('peppers.png');
B=rgb2gray(A);

C=double(B);

for i=1:size(C,1)-2
for j=1:size(C,2)-2
%Sobel mask for x-direction:
Gx=((2*C(i+2,j+1)+C(i+2,j)+C(i+2,j+2))-(2*C(i,j+1)+C(i,j)+C(i,j+2)));
%Sobel mask for y-direction:
Gy=((2*C(i+1,j+2)+C(i,j+2)+C(i+2,j+2))-(2*C(i+1,j)+C(i,j)+C(i+2,j)));

%The gradient of the image
%B(i,j)=abs(Gx)+abs(Gy);
B(i,j)=sqrt(Gx.^2+Gy.^2);

end
end
figure,imshow(B); title('Sobel gradient');

Sobel Gradient

%Define a threshold value

Thresh=100;

B=max(B,Thresh);

B(B==round(Thresh))=0;

B=uint8(B);

figure,imshow(~B);title('Edge detected Image');

Edge detected Image

Edge detected Image(Threshold value:35)

The edge detected image can be obtained from the sobel gradient by
using a threshold value.

If the sobel gradient values are lesser than the threshold value then replace it with the threshold value.
if f < threshold value then
f = threshold value.

To avoid complex computation, the gradient can also be computed using the formula:

The Image obtained from computing X-direction derivative:

The Image obtained from computing Y-direction derivative:

Also Check Sobel Edge Detection - Part 2

Like "IMAGE PROCESSING" page

Adaptive filtering-local noise filter

Adaptive filter is performed on the degraded image that contains original image and noise. The mean and variance are the two statistical measures that a local adaptive filter depends with a defined mxn window region.

MATLAB CODE:

A = imread('saturn.png');

B = rgb2gray(A);

sz = size(B,1)*size(B,2);

%Add gaussian noise with mean 0 and variance 0.005

B = imnoise(B,'gaussian',0,0.005);

figure,imshow(B); title('Image with gaussian noise');

B = double(B);

%Define the window size mxn

M = 5;

N = 5;

%Pad the matrix with zeros on all sides

C = padarray(B,[floor(M/2),floor(N/2)]);

lvar = zeros([size(B,1) size(B,2)]);

lmean = zeros([size(B,1) size(B,2)]);

temp = zeros([size(B,1) size(B,2)]);

NewImg = zeros([size(B,1) size(B,2)]);

for i = 1:size(C,1)-(M-1)

for j = 1:size(C,2)-(N-1)

temp = C(i:i+(M-1),j:j+(N-1));

tmp = temp(:);

%Find the local mean and local variance for the local region

lmean(i,j) = mean(tmp);

lvar(i,j) = mean(tmp.^2)-mean(tmp).^2;

end

%Noise variance and average of the local variance

nvar = sum(lvar(:))/sz;

%If noise_variance > local_variance then local_variance=noise_variance

lvar = max(lvar,nvar);

%Final_Image = B- (noise variance/local variance)*(B-local_mean);

NewImg = nvar./lvar;

NewImg = NewImg.*(B-lmean);

NewImg = B-NewImg;

%Convert the image to uint8 format.

NewImg = uint8(NewImg);

figure,imshow(NewImg);title('Restored Image using Adaptive Local filter');

Like "IMAGE PROCESSING" page

Add salt and pepper noise to image

How to add salt and pepper noise to an image

To obtain an image with ‘speckle’ or ‘salt and pepper’ noise we need to add white and black pixels randomly in the image matrix.

First convert the RGB image into grayscale image.

Then generate random values for the size of the matrix.

Here I used MATLAB function ‘randint’.

This function will generate random values for the given matrix size within the specified range.

For instance, consider an image matrix of size 4X3

Imgmatrix =

237 107 166

234 95 162

239 116 169

56 126 89

Generate random values for a 4X3 matrix with range 0 to 10.

randint(4,3,[0,10])

ans =

4 10 4

7 8 4

10 0 5

3 10 9

Now we can replace with pixel value zero (black) in the image matrix if there is ‘0’ value in the random matrix.

Now the image matrix will have black pixels.

Imgmatrix =

237 107 166

234 95 162

239 0 169

56 126 89

Similarly, replace the image matrix pixel value with ‘255’ if there is value ‘10’ in the random matrix.

The white pixels are now added.

Imgmatrix =

237 255 166

234 95 162

255 0 169

56 126 89

Black=0 white=255

MATLAB CODE:

A=imread('taj.jpg');

B=rgb2gray(A);

black=3;

white=253;

%Adjust the values in 'black' and 'white' to increase the noise.

NoiseImg = B;

Rmatrix = randint(size(B,1),size(B,2),[0,255]);

NoiseImg(Rmatrix <= black) = 0;

NoiseImg(Rmatrix >=white) = 255;

RImg=medfilt2(NoiseImg);

figure,subplot(1,2,1),imshow(NoiseImg),title('Add ''Salt and Pepper'' Noise');

subplot(1,2,2),imshow(RImg),title('After Noise Removal');

I used the MATLAB function 'medfilt2' to remove noise.

Learn how to remove the noise without using the 'medfilt2' function.

Like "IMAGE PROCESSING" page

I can't sleep Illusion

In this illusion, I used slanted lines in two different directions to highlight the foreground from the background.

First, the background is formed with slanting lines in one direction.

Steps to form background with slanting lines:

1. Get the size of the foreground image. For instance, MXN =5X8

2. Initialize a matrix of size 1XN=1X8.

3. Generate values for the matrix.
Eg. [10 10 250 250 10 10 250 250]ie [BBWWBBWW] B-Black, W-White

4. Initialize another matrix of size MXN. In the first row copy the random values.

5. For the second row perform circshift. By circshift the pixel at the first column will be moved to the last column and the remaining pixels are shifted to left by one.Eg [10 250 250 10 10 250 250 10].
Now the colors will be [BWWBBWWB].

6. Continue to perform circshift for all the rows. Now the slanting lines are formed.
10 10 250 250 10 10 250 250

10 250 250 10 10 250 250 10

250 250 10 10 250 250 10 10

250 10 10 250 250 10 10 250

10 10 250 250 10 10 250 250

MATLAB CODE:

%Read the Image

TImg=imread('sleep.jpg');

m=size(TImg,1);

n=size(TImg,2);

if(mod(n,2)~=0)

n=n-mod(n,2);

end

if(mod(n,4)~=0)

n=n-mod(n,4);

end

BImg=uint8(zeros([1 n 3]));

flag=0;

%Values for first column (bbwwbbwwbbww)

for i=1:2:n

if(flag==1)

BImg(1,i:i+1,:)=10;

flag=0;

else

BImg(1,i:i+1,:)=240;

flag=1;

end

%Initialization

A=uint8(zeros([m size(BImg,2) 3]));

Arot=uint8(zeros([m size(BImg,2) 3]));

SImg=uint8(zeros([m size(BImg,2) 3]));

FImg=uint8(zeros([m size(BImg,2) 3]));

A(1,:,:)=BImg(1,:,:);

%Slanting lines

for i=2:m

BImg=circshift(BImg,[1,-1]);

A(i,:,:)=BImg(1,:,:);

end

Steps to perform on foreground Image:

1. Rotate the background image from left to right.

2. Convert the foreground image into binary image and negate it. Then perform image multiplication with the rotated background image.

MATLAB CODE:

%Flip the image from left to right

Arot(:,:,1)=fliplr(A(:,:,1));

Arot(:,:,2)=fliplr(A(:,:,2));

Arot(:,:,3)=fliplr(A(:,:,3));

%Foreground

BImg=uint8(~im2bw(TImg));

BImg=imresize(BImg,[size(Arot,1) size(Arot,2)]);

SImg(:,:,1)=BImg.*Arot(:,:,1);

SImg(:,:,2)=BImg.*Arot(:,:,2);

SImg(:,:,3)=BImg.*Arot(:,:,3);

SImg(SImg==0)=1;

Image Mask:

1. Multiply the background image with the binary foreground image to obtain the mask.

MATLAB CODE:

%Background Image mask

BImg1=uint8(im2bw(TImg));

BImg1=imresize(BImg1,[size(Arot,1) size(Arot,2)]);

FImg(:,:,1)=BImg1.*A(:,:,1);

FImg(:,:,2)=BImg1.*A(:,:,2);

FImg(:,:,3)=BImg1.*A(:,:,3);

FImg(FImg==0)=1;

Combine Foreground Image and the background Image mask.

MATLAB CODE:

%Combine Background mask with foreground.

Zig=FImg.*SImg;

figure,imshow(Zig);

GUESS THE IMAGE:
If you can read this then give your reactions...

I used square font to make this illusion. You can download the font at:http://fontzone.net/font-download/SF+Square+Head+Bold/

Check this link for more optical Illusion Images: Images from Blog

Like "IMAGE PROCESSING" page

Black and White-Optical illusion

Today, Some of my friends in Facebook posted black and white optical illusion photo saying 'ITS FREAKING AWESOME'.
After seeing the image, I found it's actually black and white illusion. Yes, convert the image to binary and do logical not of the image. When you look or concentrate at one point on the image ,maybe in center you can see a logical not of the image (i.e) White pixels will be black and black pixels will be white.

The most interesting thing is, you can try with your own photo.

MATLAB CODE:

A=imread('illusion.jpg');

B=uint8(zeros(size(A)));
A=~im2bw(A);

Logical Not on Binary Image

for i=1:size(A,1)
for j=1:size(A,2)
if(A(i,j)==1)
B(i,j,:)=255;
end
end
end

Stare at the dots on image for 20 to 30 secs. Close your eyes or look up towards the ceiling.
You can see a binary image. i.e im2bw(A);

To draw the dots on the image:

C=uint8(ones(4 ,4, 3));
C(:,:,1)=C(:,:,1)*200;
C(:,:,2)=C(:,:,2)*0;
C(:,:,3)=C(:,:,3)*250;
%Find the mid point of the image and draw 3 dots.

midx=round(size(B,1)/2);
midy=round(size(B,2)/2)+30;
for i=1:size(C,1)
for j=1:size(C,2)
B(midx+i+5,midy+j,:)=C(i,j,:);
B(midx+i-5,midy+j,:)=C(i,j,:);
B(midx+i+15,midy+j,:)=C(i,j,:);
end
end

Check this Illusion also:I cant sleep Illusion
Check this link for more optical Illusion Images: Images from Blog

Like "IMAGE PROCESSING" page

Darker Edges

Captured Image

Last week my friend was in need of her signature in digital format. Fortunately, it was not for any official purpose. So we captured the signature using the camera and used it.

The image that we acquired using the camera was not that much good. So we processed the image and then included in the document.

After this thing happened, I thought of trying the image with MATLAB.

First I adjusted the image intensity using the imadjust function .
The syntax for the imadjust for rgb image is

imadjust(RGB,[Low_R Low_G Low_B; High_R High G High B],[]);

The Low values should always be lower than High values.

The range of these values between 0 to 1.

After Imadjust

Then I detected the edge of the image. Then to enlarge the image parts I dilated it. Use an appropriate structure element. Here I used 'disk'.

After that, using labeling method, I labeled all the connected components. Based on the size of the components the components where finally used and components whose size are less are ignored.

Edge detected Image

MATLAB CODE:





function my_edge

global filename A I ;

scz=get(0,'ScreenSize');

figure('Position',[round(scz(1,3)/4) round(scz(1,4)/8) 700 500],'MenuBar','None','NumberTitle','off','Name','Pencil sketch Application','Resize','off');

axes('Position',[0 0 .7 1],'xtick',[],'ytick',[]);

Low=uicontrol('Style','slider','Position',[500,280 200 20],'Max',0.99 ,'Min',0,'Value',0.0,'SliderStep',[0.05,0.1],'Callback',@draw);

High=uicontrol('Style','slider','Position',[500,370 200 20],'Max',1,'Min',0.1,'Value',.5,'SliderStep',[0.05,0.1],'Callback',@draw);

uicontrol('Style','pushbutton','String','Done','Position',[500 200 40 20],'Callback',@pushme);

directory=dir('*.jpg');

files={directory.name}';

uicontrol('style','text','Position',[500,310 100 20]','String','Low:');

uicontrol('style','text','Position',[500,395 100 20]','String','High :');

uicontrol('style','text','Position',[500,455 100 20]','String','Filename:');

uicontrol('Style','popupmenu','position',[500 420 160 30],'Value',1,'String',files,'Callback',@displayfile);



    function displayfile(obj,eve)

        ptr=get(obj,'value');

        filename=char(files(ptr));

        A=imread(filename);

        A=imresize(A,0.4);

        subplot('Position',[0.05 0.1 .6 .9]);imshow(A);

    end



    function draw(obj,eve)

        H=get(High,'Value');

        L=get(Low,'Value');

      uicontrol('Style','edit','position',[500 250 50 20],'String',L);

      uicontrol('Style','edit','position',[500 340 50 20],'String',H);

        if( L < H )


            %The intensity values for each R,G,B planes are adjusted.

            I=imadjust(A,[L L L; H H H],[]);

        end

           subplot('Position',[0.05 0.1 .6 .9]);imshow(I);

       end





                                                                                                         

                                                                                        

                                                                                         

                                                                                   

                                                                                    












































    function pushme(obj,eve)

        myedges(I);

    end

    function myedges(A)

      

A1=rgb2gray(A);



B=(edge(A(:,:,1))|edge(A(:,:,2))|edge(A(:,:,3)));

Edge detected




SE=strel('disk',3);

C=~(imdilate(B,SE));





target=(ones([size(A1,1) size(A1,2)] ));

[BW,label]=bwlabel(~C,8);



for l=1:max(max(label))

[row, col] = find(BW==l);



if(size(row,1)>30)



for i=1:size(row,1)

    x=row(i,1);

    y=col(i,1);

    target(x,y)=C(x,y);

end

end



end


subplot('Position',[0 0.5 .8 .45]);imshow(A);title('Original Image');

subplot('Position',[0 0   .8 .45]);imshow(target);title('Final Image');









    end

end

Another example:

After Imadjust and edge detection (structuring element:disk Radius=1)

Like "IMAGE PROCESSING" page