A list,

Mathematical morphology operations can be divided into binary morphology and gray morphology. Gray morphology is expanded from binary morphology. There are two basic operations of mathematical morphology, namely corrosion and expansion, and corrosion and expansion combine to form open and closed operations. Open operation is first corrosion and then expansion, closed operation is first expansion and then corrosion.

1 binary morphology

Roughly speaking, corrosion can “shrink” the target area, essentially shrinking the image’s boundaries, and can be used to eliminate small, meaningless objects. The formula is expressed as:



The formula says with structure corrosion. A, B need to be aware of is the need to define A B origin, [and the moving process of B are consistent with the process of convolution kernels mobile, with convolution kernels and image overlap again after calculation as 】 when B origin to image A translation like yuan (x, y), if B (x, y), fully contained in the image A overlapping area, (That is, all the corresponding image values of A at the position of element 1 in B are also 1) then the pixel (x,y) corresponding to the output image is assigned A value of 1, otherwise 0 is assigned.

Let’s look at a demo.



B moves on A in sequence (the same as the convolution kernel moves on the image, and then performs morphological operation on the coverage domain of B). When the coverage region of A is [1,1;1,1] or [1,0;1,1], (that is, ‘1’ in B is A subset of the coverage region), the position of the corresponding output image will be 1.

2 expansion

Roughly speaking, expansion “enlarges” the range of the target area, merging background points in contact with the target area into the target object, making the target boundary expand outwards. It can be used to fill some holes in the target area and eliminate the small particle noise contained in the target area.



In this formula, A is expanded by structure B, and the origin of structural element B is shifted to the position of image pixel (x,y). If the intersection of B and A at image pixel (x,y) is not empty (that is, at least one image value of A corresponding to the position of element 1 in B is 1), then the pixel (x,y) corresponding to the output image is assigned the value of 1; otherwise, it is assigned the value of 0.

Demo picture:



3 summary

In other words, no matter corrosion or expansion, structural element B is shifted on the image like convolution operation. The origin of structural element B is equivalent to the core center of the convolution kernel, and the result is also stored on the element at the corresponding position of the core center. However, corrosion means that B is completely contained in the area covered by it, and there is an intersection between B and the area covered by it during expansion.

Before telling the gray value morphology, we make A convention that the area of image A covered by structural element B is denoted as P (take Part meaning).

5. Corrosion of gray scale morphology

So corrosion in grayscale morphology is an operation similar to convolution. The small rectangle formed by subtracting structural element B from P can be assigned to the position of the corresponding origin by taking the minimum value.

Let’s look at an example to deepen our understanding of grayscale morphology.

Suppose we have the following image A and structural element B:



The process of gray morphological corrosion is as follows:



We specifically show the output result of the first element of the output image, that is, the position of 4 corresponding to the origin. The values of the other elements of the output image are also obtained in this way. We’ll see that the region that B first covers is the subtracted matrix, and then we take min in the difference matrix to be the value of the origin.



Expansion of grayscale morphology

According to the above description of corrosion, expansion is described in the same way. Expansion in grayscale morphology is an operation similar to convolution. P is added to B, and then the maximum value in this region is assigned to the position corresponding to the origin of structural element B.





Here is also the origin of the value of the first element of the output image.



The maximum value of the sum of the above matrices is 6, so assign 6 to the position corresponding to the origin of the structural element.

6 Summary Above introduced the concept of gray morphology, here to say their use. Compared with the original image, the corrosion results in smaller pixels than before, so it is suitable for removing peak noise. As a result of gray expansion, each pixel becomes larger than the previous one, so it is suitable for removing the trough noise.

Ii. Source code

function varargout = untitled(varargin)
% UNTITLED M-file for untitled.fig
%      UNTITLED, by itself, creates a new UNTITLED or raises the existing
%      singleton*.
%
%      H = UNTITLED returns the handle to a new UNTITLED or the handle to
%      the existing singleton*.
%
%      UNTITLED('CALLBACK',hObject,eventData,handles,...) calls the local
%      function named CALLBACK in UNTITLED.M with the given input arguments.
%
%      UNTITLED('Property'.'Value',...). creates anew UNTITLED or raises the
%      existing singleton*.  Starting from the left, property value pairs are
%      applied to the GUI before untitled_OpeningFcn gets called.  An
%      unrecognized property name orinvalid value makes property application % stop. All inputs are passed to untitled_OpeningFcn via varargin. % % *See GUI  Options on GUIDE's Tools menu.  Choose "GUI allows only one % instance to run (singleton)".
%
% See also: GUIDE, GUIDATA, GUIHANDLES

% Edit the above text to modify the response to help untitled

% Last Modified by GUIDE v2. 5 21-May- 2021. 23:07:23

% Begin initialization code - DO NOT EDIT

gui_Singleton = 1;
gui_State = struct('gui_Name',       mfilename, ...
                   'gui_Singleton',  gui_Singleton, ...
                   'gui_OpeningFcn', @untitled_OpeningFcn, ...
                   'gui_OutputFcn',  @untitled_OutputFcn, ...
                   'gui_LayoutFcn', [],...'gui_Callback'[]);if nargin && ischar(varargin{1})
    gui_State.gui_Callback = str2func(varargin{1});
end

if nargout
    [varargout{1:nargout}] = gui_mainfcn(gui_State, varargin{:});
else
    gui_mainfcn(gui_State, varargin{:});
end
% End initialization code - DO NOT EDIT


% --- Executes just before untitled is made visible.
function untitled_OpeningFcn(hObject, eventdata, handles, varargin)
% This function has no output args, see OutputFcn.
% hObject    handle to figure
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)
% varargin   command line arguments to untitled (see VARARGIN)

% Choose default command line output for untitled
handles.output = hObject;

% Update handles structure
guidata(hObject, handles);

% UIWAIT makes untitled wait for user response (see UIRESUME)
% uiwait(handles.figure1);


% --- Outputs from this function are returned to the command line.
function varargout = untitled_OutputFcn(hObject, eventdata, handles) 
% varargout  cell array for returning output args (see VARARGOUT);
% hObject    handle to figure
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)

% Get default command line output from handles structure
varargout{1} = handles.output;


% --- Executes on button press in pushbutton1.
function pushbutton1_Callback(hObject, eventdata, handles)
% hObject    handle to pushbutton1 (see GCBO)
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)

global im;
global str;
[filename,pathname]=uigetfile_new({'*. *'},'Select training picture... ');
str=[pathname  filename];
im=imread(str);
axes(handles.axes1);
imshow(im);



% --- Executes on button press in pushbutton2.
function pushbutton2_Callback(hObject, eventdata, handles)
% hObject    handle to pushbutton2 (see GCBO)
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)
global im1;
global str1;
[filename,pathname]=uigetfile_new({'*. *'},'Select test picture... ');
str1=[pathname  filename];
im1=imread(str1);
axes(handles.axes2);
imshow(im1);


% --- Executes on button press in pushbutton3.
function pushbutton3_Callback(hObject, eventdata, handles)
% hObject    handle to pushbutton3 (see GCBO)
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)
global im;
global im1;
global str1;
se1 = strel('disk'.5);
se2 = strel('square'.16);
thresh1 = 0.2;
thresh2 = 0.01;
imtra = im2double(im);
rt = imtra(:,:,1);
gt = imtra(:,:,2);
bt = imtra(:,:,3);
idxr1 = find(rt>0);
idxg1 = find(gt>0);
idxb1 = find(bt>0);

mr1 = mean(rt(idxr1));
mg1 = mean(gt(idxg1));
mb1 = mean(bt(idxb1));

clear idxr1 idxg1 idxb1 idxgr1;
im1=imread(str1);
I1 = im2double(im1);
col = size(I1,2);
rate = 1600/col;
I1 = imresize(I1,rate);
r1 = I1(:,:,1);
g1 = I1(:,:,2);
b1 = I1(:,:,3);
[row1,col1] = size(r1);
r1 = abs(r1 - mr1);
g1 = abs(g1 - mg1);
b1 = abs(b1 - mb1);


x1 = (r1+g1+b1)./3;
figure(1);
imshow(x1,[]);
title(Minus the color component);
x1 = im2bw(x1,10/255);

figure(12);
imshow(x1,[]);
title('Binary image');

x1 = imopen(x1,se1);
x1 = imclose(x1,se2);
figure(2);
imshow(x1,[]);
title('After the open and close operation');
y1 = adjvar(g1);
y1 = im2bw(y1,50/255);
figure(3);
imshow(y1,[]);
title('Calculate pixel variance and binarize');
x1 = -x1+1; x1_labeled = bwlabel(x1); X1_num = Max (Max (x1_labeled)); % displays the original image for later marking figure in the loop (4);
imshow(I1);
title('Recognition result');
for i = 1:x1_num
    x1_idx = find(x1_labeled==i);
    y1_idx = find(y1(x1_idx)==1); % If the ratio of foreground pixel number to total pixel number of connected region in the current connected region corresponding to the binary image of variance is greater than a certain threshold value of %, it is considered to be Maltang grass and marked on the original image. The position to be marked is not accurately calculated here.if(size(y1_idx,1)/size(x1_idx,1)>thresh1)
        idx = x1_idx(floor(1+size(x1_idx,1) /2));
        r = mod(idx,row1);
        c = ceil(idx/row1);
       text(c,r,'Horseweed'.'color'.'black');
    end
end



% --- Executes on button press in pushbutton4.
function pushbutton4_Callback(hObject, eventdata, handles)
% hObject    handle to pushbutton4 (see GCBO)
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)
clear all
clc
close(gcf)
Copy the code

3. Operation results

















Fourth, note

Version: 2014 a