I need the MATLAB coding on how to do the questions above ( C,D,E, and F) using
ID: 3577380 • Letter: I
Question
I need the MATLAB coding on how to do the questions above (C,D,E, and F) using MATLAB. Questions A and B has been answer, see coding below. Thank you
% ---------------- Step 1 ------------------------
% Read in an image, get information
% type help imread for more information
InputImage = 'IDPicture.bmp';
%OutputImage1 = 'IDPicture_bw.bmp';
C1 = imread(InputImage);
[ROWS COLS CHANNELS] = size(C1);
% ---------------- Step 2 ------------------------
% If you want to display the three separate bands
% with the color image in one window, here is
% what you need to do
% Basically you generate three "color" images
% using the three bands respectively
% and then use [] operator to concatenate the four images
% the orignal color, R band, G band and B band
% First, generate a blank image. Using "uinit8" will
% give you an image of 8 bits for each pixel in each channel
% Since the Matlab will generate everything as double by default
CR1 =uint8(zeros(ROWS, COLS, CHANNELS));
% Note how to put the Red band of the color image C1 into
% each band of the three-band grayscale image CR1
for band = 1 : CHANNELS,
CR1(:,:,band) = (C1(:,:,1));
end
% Do the same thing for G
CG1 =uint8(zeros(ROWS, COLS, CHANNELS));
for band = 1 : CHANNELS,
CG1(:,:,band) = (C1(:,:,2));
end
% and for B
CB1 =uint8(zeros(ROWS, COLS, CHANNELS));
for band = 1 : CHANNELS,
CB1(:,:,band) = (C1(:,:,3));
end
% Whenever you use figure, you generate a new figure window
No1 = figure; % Figure No. 1
%This is what I mean by concatenation
disimg = [C1, CR1;CG1, CB1];
% Then "image" will do the display for you!
image(disimg);
% ---------------- Step 3 ------------------------
% Now we can calculate its intensity image from
% the color image. Don't forget to use "uint8" to
% covert the double results to unsigned 8-bit integers
I1 = uint8(round(sum(C1,3)/3));
% You can definitely display the black-white (grayscale)
% image directly without turn it into a three-band thing,
% which is a waste of memeory space
No2 = figure; % Figure No. 2
image(I1);
% If you just stop your program here, you will see a
% false color image since the system need a colormap to
% display a 8-bit image correctly.
% The above display uses a default color map
% which is not correct. It is beautiful, though
% ---------------- Step 4 ------------------------
% So we need to generate a color map for the grayscale
% I think Matlab should have a function to do this,
% but I am going to do it myself anyway.
% Colormap is a 256 entry table, each index has three entries
% indicating the three color components of the index
MAP =zeros(256, 3);
% For a gray scale C[i] = (i, i, i)
% But Matlab use color value from 0 to 1
% so I scale 0-255 into 0-1 (and note
% that I do not use "unit8" for MAP
for i = 1 : 256, % a comma means pause
for band = 1:CHANNELS,
MAP(i,band) = (i-1)/255;
end
end
%call colormap to enfore the MAP
colormap(MAP);
% I forgot to mention one thing: the index of Matlab starts from
% 1 instead 0.
% Is it correct this time? Remember the color table is
% enforced for the current one, which is the one we
% just displayed.
% You can test if I am right by try to display the
% intensity image again:
No3 = figure; % Figure No. 3
image(I1);
(1) (30 points) Image formation. In this small project, you are going to use Matlab to read, manipulate and write image data. The purpose of the project is to make you familiar with the basic digital image formations. Your program should do the following things: a. Read in color intage Ci(Ky) (R(xy),G(xy),B(xy) in Windows BMP format, and display it b. Display the images of the three color components,R(xy,G(xy) and B(xy), separately. You should display three black-white-like images c. Generate an intensity image IOK,y) and display it. You should use the equation I-0.299R 0.587G 4B (the NTSC standard for luminance) d. The original intensity image should have 256 gray levels. Please uniformly quantize this image into K levels (with K-4, 16, 32, 64). As an example, when K 2 pixels whose values are below 128 are turned to0, otherwise to 255. Display the four quantized images with four different K levels and tell us how the images still look like the original ones e. Quantize the original three-band color image ci(x,y) into K level color images CK(xy (R xy. xy)) (with uniform intervals), and display them. You may choose K 2 and 4 for each band). Do they have any advantages in viewing and/or in computer processing (e.g. segmentation? f. antize the original three-band color image C1(xy) into a color image CL(xy) (R (xy),G x,y), B xy) th a logarithmic function) and display it. You may choose a function l' C In (I+1) (for each band), where I is the original value (0-255) I' s the quantized value, and Cis a constant to scale I nto (0-255), and ln is the natural logarithm. Please find the best C value so for an input in the range of 0-255, the output range is still 0 255. Note that when I 0,I 0 too Please for each of the above, provide your observations conclusions, rather than just show the experimental results.Explanation / Answer
% ---------------- Step 1 ------------------------
% Read in an image, get information
% type help imread for more information
InputImage = 'IDPicture.bmp';
%OutputImage1 = 'IDPicture_bw.bmp';
C1 = imread(InputImage);
[ROWS COLS CHANNELS] = size(C1);
% ---------------- Step 2 ------------------------
% If you want to display the three separate bands
% with the color image in one window, here is
% what you need to do
% Basically you generate three "color" images
% using the three bands respectively
% and then use [] operator to concatenate the four images
% the orignal color, R band, G band and B band
% First, generate a blank image. Using "uinit8" will
% give you an image of 8 bits for each pixel in each channel
% Since the Matlab will generate everything as double by default
CR1 =uint8(zeros(ROWS, COLS, CHANNELS));
% Note how to put the Red band of the color image C1 into
% each band of the three-band grayscale image CR1
for band = 1 : CHANNELS,
CR1(:,:,band) = (C1(:,:,1));
end
% Do the same thing for G
CG1 =uint8(zeros(ROWS, COLS, CHANNELS));
for band = 1 : CHANNELS,
CG1(:,:,band) = (C1(:,:,2));
end
% and for B
CB1 =uint8(zeros(ROWS, COLS, CHANNELS));
for band = 1 : CHANNELS,
CB1(:,:,band) = (C1(:,:,3));
end
% Whenever you use figure, you generate a new figure window
No1 = figure; % Figure No. 1
%This is what I mean by concatenation
disimg = [C1, CR1;CG1, CB1];
% Then "image" will do the display for you!
image(disimg);
% ---------------- Step 3 ------------------------
% Now we can calculate its intensity image from
% the color image. Don't forget to use "uint8" to
% covert the double results to unsigned 8-bit integers
I1 = uint8(round(sum(C1,3)/3));
% You can definitely display the black-white (grayscale)
% image directly without turn it into a three-band thing,
% which is a waste of memeory space
No2 = figure; % Figure No. 2
image(I1);
% If you just stop your program here, you will see a
% false color image since the system need a colormap to
% display a 8-bit image correctly.
% The above display uses a default color map
% which is not correct. It is beautiful, though
% ---------------- Step 4 ------------------------
% So we need to generate a color map for the grayscale
% I think Matlab should have a function to do this,
% but I am going to do it myself anyway.
% Colormap is a 256 entry table, each index has three entries
% indicating the three color components of the index
MAP =zeros(256, 3);
% For a gray scale C[i] = (i, i, i)
% But Matlab use color value from 0 to 1
% so I scale 0-255 into 0-1 (and note
% that I do not use "unit8" for MAP
for i = 1 : 256, % a comma means pause
for band = 1:CHANNELS,
MAP(i,band) = (i-1)/255;
end
end
%call colormap to enfore the MAP
colormap(MAP);
% I forgot to mention one thing: the index of Matlab starts from
% 1 instead 0.
% Is it correct this time? Remember the color table is
% enforced for the current one, which is the one we
% just displayed.
% You can test if I am right by try to display the
% intensity image again:
No3 = figure; % Figure No. 3
image(I1);
% See???
% You can actually check the color map using
% the edit menu of each figure window
% ---------------- Step 5 ------------------------
% Use imwrite save any image
% check out image formats supported by Matlab
% by typing "help imwrite
% imwrite(I1, OutputImage1, 'BMP');
% ---------------- Step 6 and ... ------------------------
% Students need to do the rest of the jobs from c to g.
% Write code and comments - turn it in both in hard copies and
% soft copies (electronically)
% ---------------------------------------------------------
% ---------------- PART 1C --------------------------------
% ---------------------------------------------------------
% Generate an intensity image I(x,y) and display it.
% You should use the equation
% I = 0.299R + 0.587G + 0.114B (the NTSC standard for luminance).
IntensityImage = (0.299 * CR1) + (0.587 * CG1) + (0.114 * CB1);
No4 = figure;
image(IntensityImage);
title('Intensity Image - I = 0.299R + 0.587G + 0.114B')
% ---------------------------------------------------------
% ---------------- PART 1D --------------------------------
% ---------------------------------------------------------
% Uniformly quantize this image into K levels ( with K=4, 16, 32, 64).
No5 = figure;
K = [4 16 32 64];
for i = 1:length(K)
I_thres = IntensityImage;
intervals = round(linspace(1,256,K(i)));
for j = 1:length(intervals)-1
I_thres(I_thres > intervals(j) & I_thres < intervals(j+1)) = intervals(j);
end
subplot(2, 2, i);
image(I_thres);
t = strcat('K = ', num2str(K(i)));
title(t);
end
% ---------------------------------------------------------
% ---------------- PART 1E --------------------------------
% ---------------------------------------------------------
No6 = figure;
K = [2 4];
for i = 1:length(K)
Org_image_a = C1;
Rband = Org_image_a(:, :, 1);
Gband = Org_image_a(:, :, 2);
Bband = Org_image_a(:, :, 3);
y = linspace(1, 256, K(i) + 1);
intervals = round(linspace(0, 256, K(i)))
for j = 1:numel(intervals)-1
Rband(Rband > y(j) & Rband < y(j+1)) = intervals(j);
Gband(Gband > y(j) & Gband < y(j+1)) = intervals(j);
Bband(Bband > y(j) & Bband < y(j+1)) = intervals(j);
end
Org_image_a = cat(3, Rband, Gband, Bband);
subplot(1, 2, i);
image(Org_image_a);
t = strcat('K = ', num2str(K(i)));
title(t);
end
% ---------------------------------------------------------
% ---------------- PART 1F --------------------------------
% ---------------------------------------------------------
Org_image_1 = C1;
Rband_1_1 = Org_image_1(:, :, 1);
Gband_1_1 = Org_image_1(:, :, 2);
Bband_1_1 = Org_image_1(:, :, 3);
c = 0.15;
Rband_1_ = double(Rband_1_1);
Bband_1_ = double(Bband_1_1);
Gband_1_ = double(Gband_1_1);
Rband_1 = c.*log(1+Rband_1_);
Bband_1 = c.*log(1+Bband_1_);
Gband_1 = c.*log(1+Gband_1_);
Org_image_1 = cat(3, Rband_1, Gband_1, Bband_1);
figure
subplot(2, 4, 1);
imshow(C1);
title('Original Image');
subplot(2, 4, 2);
imshow(Org_image_1);
title('Log Transformation Applied');
subplot(2, 4, 3);
imshow(Rband_1_1);
title('Red Band Org');
subplot(2, 4, 4);
imshow(Rband_1);
title('Red Log Band');
subplot(2, 4, 5);
imshow(Gband_1_1);
title('Green Band Org');
subplot(2, 4, 6);
imshow(Gband_1);
title('Green Log Band');
subplot(2, 4, 7);
imshow(Bband_1_1);
title('Blue Band Org');
subplot(2, 4, 8);
imshow(Bband_1);
title('Blue Log Band');
% % ---------------------------------------------------------
% % ---------------- PART 2 --------------------------------
% % ---------------------------------------------------------
% Histogram Equalization
I = imread('IDpicture.bmp');
J = rgb2gray(I);
size(I)
new_img = HistEq(J);
figure, image(I);
figure, image(new_img);
title('Histogram equalization');
% Histogram of the Equalized Image
No8 = figure;
img_hist = Histogram(new_img);
x = linspace(1, 256, 256);
bar(x, img_hist, 'FaceColor', [0 .5 .5], 'EdgeColor', [0 .9 .9], 'LineWidth', 1.5);
set(gca,'XLim',[1 256]);
title('Histogram of the Original Image')
% Thresholding
No9 = figure;
Test_img = IntensityImage;
img_hist = Histogram(Test_img);
[row, col] = size(IntensityImage);
img_size = row * col;
threshold = OtsuThreshold(img_hist, img_size);
Test_img(Test_img > threshold) = 256;
Test_img(Test_img < threshold) = 0;
image(Test_img);
title('Otsu Threshold Applied');
%
% % ---------------------------------------------------------
% % ---------------- PART 3 --------------------------------
% % ---------------------------------------------------------
% %
% 1x2 operator
x12 = imread('IDpicture.bmp');
x12_gray = rgb2gray(x12);
x12_gray_d = double(x12_gray);
[x12_row, x12_col] = size(x12_gray_d);
x12_thres = 23;
for i=1:x12_col
for j=1:x12_row
if (i == 1 || j == 1 || i == x12_col || j == x12_row)
x12_image(j, i) = 255;
else
sx = (-(x12_gray_d(j, i)) + x12_gray_d(j + 1, i));
sy = (-(x12_gray_d(j, i)) + x12_gray_d(j, i + 1));
yM = sqrt(sy.^2);
xM = sqrt(sx.^2);
M = sqrt(sx.^2 + sy.^2);
x12grad(j, i) = M;
if (M < x12_thres)
x12_image(j, i) = 0;
else
x12_image(j, i) = 255;
end
if (yM > x12_thres)
my_y_image_x12(j, i) = 0;
else
my_y_image_x12(j, i) = 255;
end
if (xM > x12_thres)
my_x_image_x12(j, i) = 0;
else
my_x_image_x12(j, i) = 255;
end
end
end
end
figure
subplot(2, 2, 1);
imshow(x12);
title('Original Image')
subplot(2, 2, 2);
imshow(my_y_image_x12);
title('horizontal gradients')
subplot(2, 2, 3);
imshow(my_x_image_x12);
title('vertical gradients')
subplot(2, 2, 4);
imshow(x12_image);
title('combined gradients')
figure,imshow(x12grad);
title('1x2 gradient');
% sobel operator
my_image = imread('IDpicture.bmp');
my_img = rgb2gray(my_image);
grad = my_img;
my_img = double(my_img);
[my_row, my_col] = size(my_img);
my_threshold = 107;
for i=1:my_col
for j=1:my_row
if (i == 1 || j == 1 || i == my_col || j == my_row)
my_image_sobel(j, i) = 255;
my_x_image_sobel(j, i) = 255;
my_y_image_sobel(j, i) = 255;
else
sx = -(my_img(j - 1, i - 1) + 2 * my_img(j, i - 1) + my_img(j + 1, j - 1)) + my_img(j - 1, i + 1) + 2 * my_img(j, i + 1) + my_img(j + 1, i + 1);
sy = -(my_img(j - 1, i - 1) + 2 * my_img(j - 1, i) + my_img(j - 1, i + 1)) + my_img(j + 1, i - 1) + 2 * my_img(j + 1, i) + my_img(j + 1, i + 1);
M = sqrt(sx.^2 + sy.^2);
yM = sqrt(sy.^2);
xM = sqrt(sx.^2);
grad(j, i) = M;
if (M > my_threshold)
my_image_sobel(j, i) = 0;
else
my_image_sobel(j, i) = 255;
end
if (yM > my_threshold)
my_y_image_sobel(j, i) = 0;
else
my_y_image_sobel(j, i) = 255;
end
if (xM > my_threshold)
my_x_image_sobel(j, i) = 0;
else
my_x_image_sobel(j, i) = 255;
end
end
end
end
figure
subplot(2, 2, 1);
imshow(my_image);
title('Original Image')
subplot(2, 2, 2);
imshow(my_y_image_sobel);
title('horizontal gradients')
subplot(2, 2, 3);
imshow(my_x_image_sobel);
title('vertical gradients')
subplot(2, 2, 4);
imshow(my_image_sobel);
title('combined gradients')
figure,imshow(grad);
title('Sobel gradient');
figure;
img_hist = Histogram(grad);
x = linspace(1, 256, 256);
bar(x, img_hist, 'FaceColor', [0 .5 .5], 'EdgeColor', [0 .9 .9], 'LineWidth', 1.5);
set(gca,'XLim',[1 256]);
title('Histogram of the Gradient Image')
cdf = cumsum(my_image_sobel) / sum(my_image_sobel);
index95 = find(cdf >= 0.65, 1, 'first');
grad(grad < index95) = 0;
figure,imshow(grad);
title('Sobel Edges');