教程1：Matlab图像处理基础（一）
本周主要任务：完成下面作业
作业1： 逐行在Matlab里面实现下列代码
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Introduction to Matlab
% (adapted from https://cs.brown.edu/courses/cs143/docs/matlab-tutorial/)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Symbol "%"is used to comment a line
%% Helpful commands
clc;
clear all;
close all;
%% Basics
% data type
5 % integer 32-bit
3.1415 % double 64-bit
isfloat(5.0)
% Arithmetic
1 + 2
1 / 2
2 ^ 5
% Logic
1 == 0
1 ~= 0
1 && 0
1 || 0
% variable
a = 4 % displays output
b = 0.5; % ";" suppresses output
c = a ^ b
%% Vector and matrix, Basic
% vector
s = 5 % A scalar
r = [1 2 3] % A row vector
c = [4; 5; 6] % A column vector
v = r'
v = 1:3 % from 1 to 3, with a default size of 1
v = 1:.5:3 % from 1 to 3, with a step size of 0.5
% matrix
A = [1 2 3;4 5 6] % 2x3 matrix, ";" separates row, "," separates column (optional)
B = zeros(2,4) % 2x4 matrix filled with zero
B = ones(2,4) % 2x4 matrix filled with one
C = eye(3) % 3x3 identity matrix
% indexing
v= 1:.5:3
v(1) % vector element
A = [1 2 3 4; 5 6 7 8; 9 10 11 12]
A(2,3) % matrix element
A(2,:) % 2nd row
A(2:end,2) % [6;10;14]
size(A)
size(A, 1) % number of rows
size(A, 2) % number of columns
numel(A) % number of elements
%% Vector and matrix, operations
% vector
a = 1:3
b = 4:6
a + b
2 * a % scalar multiplication
a .* b % element-wise multiplication
a * b' % vector multiplication
% built-in functions
sum(a) % Sum of vector elements
mean(a) % Mean of vector elements
var(a) % Variance of elements
std(a) % Standard deviation
max(a) % Maximum
min(a) % Minimum
log(a) % Element-wise logarithm
% matrix
A = [1 2; 3 4; 5 6]; % A 3x2 matrix
B = [5 6 7]; % A 1x3 row vector
B * A % Vector-matrix product results in
% a 1x2 row vector
B = A(:) % 6x1 column vector
max(B(:)) % max of all elements
A = reshape(B, 2, 3)
C = repmat(B, 1, 2)
[U, S, V] = svd(A)
%% Control flow
% Copy positive elements from A to B
% Implementation using loops:
A = [1 -1 1; -1 1 -1; 1 -1 1]
[m,n] = size(A);
B = zeros(m,n);
for i=1:m
for j=1:n
if A(i,j)>0
B(i,j) = A(i,j);
end
end
end
B
% All this can be computed w/o any loop!
B = zeros(m,n);
ind = find(A > 0); % Find indices of positive elements of A
% (see "help find" for more info)
B(ind) = A(ind); % Copies into B only the elements of A
% that are > 0
B
%% Image Processing
img = imread('superman.png');
imshow(img);
MINGREEN = 173; MAXGREEN = 173;
mask = ~(MINGREEN <= img(:,:,2) & img(:,:,2) <= MAXGREEN);
img_new = repmat(mask,[1 1 3]) .* double(img);
figure;
imshow(uint8(img_new));
%% Plot
figure; % Open new figure
x = pi*[-24:24]/24; % 1x49 row vec
plot(x, sin(x));
xlabel('radians'); % Assign label for x-axis
ylabel('sin value'); % Assign label for y-axis
title('Sin wave'); % Assign plot title
作业2：(superman, space的图片附在后面)
vectorize = false;
% Read in the images. They must be the same size.
superman = imread('superman.png');
space = imread('space.png');
characterImg = superman;
backgroundImg = space;
% Create a matrix matching the dimensions of characterImg.
outputImg = zeros(size(characterImg));
height = size(characterImg, 1);
width = size(characterImg, 2);
% By trial and error, I determined that the green screen pixels are within
% these color values.
MINRED = 10; MINGREEN = 100; MINBLUE = 10;
MAXRED = 160; MAXGREEN = 220; MAXBLUE = 110;
if ~vectorize
% The straightforward implementation using nested for loops.
for y = 1:height
for x = 1:width
redMatch = MINRED <= superman(y, x, 1) && superman(y, x, 1) <= MAXRED;
greenMatch = MINGREEN <= superman(y, x, 2) && superman(y, x, 2) <= MAXGREEN;
blueMatch = MINBLUE <= superman(y, x, 3) && superman(y, x, 3) <= MAXBLUE;
match = redMatch && greenMatch && blueMatch;
if match
outputImg(y,x,:) = backgroundImg(y,x,:);
else
outputImg(y,x,:) = characterImg(y,x,:);
end
end
end
else
% The advanced, fast implementation using matrix math.
% Make a 2D array that holds a '1' where red is within bounds.
redMatchImg = (MINRED <= superman(:,:,1)) & (superman(:,:,1) <= MAXRED);
greenMatchImg = MINGREEN <= superman(:,:,2) & superman(:,:,2) <= MAXGREEN;
blueMatchImg = MINBLUE <= superman(:,:,3) & superman(:,:,3) <= MAXBLUE;
% Holds a '1' where all colors matched the green screen.
matchImg = redMatchImg & greenMatchImg & blueMatchImg;
% Duplicate the matrix to convert it from 480x640 to 480x640x3.
% This makes it easy to do an element-by-element multiplication with the images.
matchImg = repmat(matchImg, [1 1 3]);
% Make another matrix which is '1' everywhere the original was '0', and vice versa
nonMatchImg = ~matchImg;
% Now, multiply the character image by '1' where there's no green screen, and
% multiply the background by '1' where there IS green screen. This
% replaces the "if" statement above.
outputImg = double(characterImg) .* nonMatchImg + double(backgroundImg) .* matchImg;
% The cast to double is to deal with a MATLAB quirk: images are read in
% as arrays of type "uint8" (a.k.a. byte). But MATLAB generally needs to work in
% doubles. So we'll convert to double to do math, and then convert back
% to uint8, below, to display.
end
% Note that, if the image color format is 0 to 255, we need to cast to uint8 for it to display.
imshow(uint8(outputImg), []);
space.jpg图片如下：
superman.jpg 图片如下：
