手写数字识别基于支持向量机SVM实现手写数字识别matlab源码含GUI

Posted Matlab咨询QQ1575304183

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了手写数字识别基于支持向量机SVM实现手写数字识别matlab源码含GUI相关的知识,希望对你有一定的参考价值。

一、简介

支持向量机(Support Vector Machine)是Cortes和Vapnik于1995年首先提出的,它在解决小样本、非线性及高维模式识别中表现出许多特有的优势,并能够推广应用到函数拟合等其他机器学习问题中。
1 数学部分
1.1 二维空间
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
2 算法部分
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
在这里插入图片描述

二、源代码

function varargout = DigitClassifyUI(varargin)

%


% DIGITCLASSIFYUI MATLAB code for DigitClassifyUI.fig

% DIGITCLASSIFYUI, by itself, creates a new DIGITCLASSIFYUI or raises the existing

% singleton*.

%

% H = DIGITCLASSIFYUI returns the handle to a new DIGITCLASSIFYUI or the handle to

% the existing singleton*.

%

% DIGITCLASSIFYUI('CALLBACK',hObject,eventData,handles,...) calls the local

% function named CALLBACK in DIGITCLASSIFYUI.M with the given input arguments.

%

% DIGITCLASSIFYUI('Property','Value',...) creates a new DIGITCLASSIFYUI or raises the

% existing singleton*. Starting from the left, property value pairs are

% applied to the GUI before DigitClassifyUI_OpeningFcn gets called. An

% unrecognized property name or invalid value makes property application

% stop. All inputs are passed to DigitClassifyUI_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 DigitClassifyUI


% Last Modified by GUIDE v2.5 10-Feb-2021 18:44:08


% Begin initialization code - DO NOT EDIT

gui_Singleton = 1;

gui_State = struct('gui_Name', mfilename, ...

'gui_Singleton', gui_Singleton, ...

'gui_OpeningFcn', @DigitClassifyUI_OpeningFcn, ...

'gui_OutputFcn', @DigitClassifyUI_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 DigitClassifyUI is made visible.

function DigitClassifyUI_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 DigitClassifyUI (see VARARGIN)


% Choose default command line output for DigitClassifyUI

handles.output = hObject;


% Update handles structure

guidata(hObject, handles);


% UIWAIT makes DigitClassifyUI wait for user response (see UIRESUME)

% uiwait(handles.figure1);

global FigHandle AxesHandle RectHandle;

FigHandle = handles.output;

AxesHandle = handles.axes_write;

MouseDraw();

axis(handles.axes_write,[1 400 1 400]); % 设定图轴范围

RectHandle = rectangle(handles.axes_write,'Position',[80,66,240,268],'LineStyle','--','EdgeColor','#a9a9a9');


% --- Outputs from this function are returned to the command line.

function varargout = DigitClassifyUI_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)



% --- Executes on button press in pushbutton_loadImage.

function pushbutton_loadImage_Callback(hObject, eventdata, handles)

% hObject handle to pushbutton_loadImage (see GCBO)

% eventdata reserved - to be defined in a future version of MATLAB

% handles structure with handles and user data (see GUIDATA)

global RectHandle;

cla(handles.axes_write, 'reset')

set(handles.axes_write, 'Visible','off');


set(handles.output, 'Pointer', 'arrow');

axis(handles.axes_write,[1 400 1 400]); % 设定图轴范围

RectHandle = rectangle(handles.axes_write,'Position',[80,66,240,268],'LineStyle','--','EdgeColor','#a9a9a9');


% 弹出文件选择框,选择一张图片

[file,path] = uigetfile({'*.jpg;*.jpeg;*.png;*.bmp;*.tif',...

'图片文件 (*.jpg,*.jpeg,*.png,*.bmp,*.tif)'},'选择一张图片');

if isequal(file,0) % 若文件不存在

set(handles.edit_imagePath, 'String','请选择一张图片');

else

fileName= fullfile(path, file); % 选择的图片绝对路径

set(handles.edit_imagePath, 'String', fileName); % 显示选择的图片路径

InputImage = imread(fileName);

image(handles.axes_raw, InputImage);

set(handles.axes_raw, 'Visible','off');


set(gcf, 'Pointer', 'arrow');

set(gcf, 'WindowButtonMotionFcn', '')

set(gcf, 'WindowButtonUpFcn', '')



% 开始执行预处理

if numel(size(InputImage))==3

InputImage = rgb2gray(InputImage); % 灰度化图片

axes(handles.axes_gray);

imshow(InputImage);

else

axes(handles.axes_gray);

imshow(InputImage);

end

% 二值化

InputImage = imbinarize(InputImage);

axes(handles.axes_binary);

imshow(InputImage);


% 特征提取

InputImage = imresize(InputImage, [28, 28]);

cellSize = [4 4];

[~, vis4x4] = extractHOGFeatures(InputImage,'CellSize',[4 4]);

axes(handles.axes_features);

plot(vis4x4);


load('trainedSvmModel.mat','classifier');

features(1, :) = extractHOGFeatures(InputImage,'CellSize',cellSize);

predictedLabel = predict(classifier, features);

str = string(predictedLabel);

set(handles.text_result, 'String', str);

end

axes(handles.axes_write);

MouseDraw();

% set(gcf, 'WindowButtonDownFcn', '');




% --- Executes on button press in pushbutton_load.

function pushbutton_load_Callback(hObject, eventdata, handles)

% hObject handle to pushbutton_load (see GCBO)

% eventdata reserved - to be defined in a future version of MATLAB

% handles structure with handles and user data (see GUIDATA)

global RectHandle;

axis(handles.axes_write,[1 400 1 400]); % 设定图轴范围

set(handles.edit_imagePath, 'String','请选择一张图片');

delete(RectHandle);

h=getframe(handles.axes_write);

imwrite(h.cdata,'writedImage.jpg');


InputImage = imread('writedImage.jpg');

% InputImage = cat(3, InputImage,InputImage,InputImage);

image(handles.axes_raw,InputImage);

set(handles.axes_raw, 'Visible','off');

axis(handles.axes_write,[1 400 1 400]); % 设定图轴范围

RectHandle = rectangle(handles.axes_write,'Position',[80,66,240,268],'LineStyle','--','EdgeColor','#a9a9a9');

global FigHandle

set(FigHandle, 'Pointer', 'arrow');

set(FigHandle, 'WindowButtonMotionFcn', '')

set(FigHandle, 'WindowButtonUpFcn', '')

set(FigHandle, 'WindowButtonDownFcn', '');


% 开始执行预处理

if numel(size(InputImage))==3

InputImage = rgb2gray(InputImage); % 灰度化图片

axes(handles.axes_gray);

imshow(InputImage);

else

axes(handles.axes_gray);

imshow(InputImage);

end

% 二值化

InputImage = imbinarize(InputImage);

axes(handles.axes_binary);

imshow(InputImage);


% 特征提取

InputImage = imresize(InputImage, [28, 28]);

cellSize = [4 4];

[~, vis4x4] = extractHOGFeatures(InputImage,'CellSize',[4 4]);

axes(handles.axes_features);

plot(vis4x4);


load('trainedSvmModel.mat','classifier');

features(1, :) = extractHOGFeatures(InputImage,'CellSize',cellSize);

predictedLabel = predict(classifier, features);

str = string(predictedLabel);

set(handles.text_result, 'String', str);

MouseDraw();


% --- Executes on button press in pushbutton_clear.

function pushbutton_clear_Callback(hObject, eventdata, handles)

% hObject handle to pushbutton_clear (see GCBO)

% eventdata reserved - to be defined in a future version of MATLAB

% handles structure with handles and user data (see GUIDATA)

global RectHandle;

global FigHandle

set(FigHandle, 'Pointer', 'arrow');

set(FigHandle, 'WindowButtonMotionFcn', '')

set(FigHandle, 'WindowButtonUpFcn', '')

set(FigHandle, 'WindowButtonDownFcn', '');

set(handles.edit_imagePath, 'String','请选择一张图片');

set(handles.text_result, 'String', 'None');

cla(handles.axes_write, 'reset')

set(handles.axes_write, 'Visible','off');

cla(handles.axes_raw, 'reset')

set(handles.axes_raw, 'Visible','off');

cla(handles.axes_gray, 'reset')

set(handles.axes_gray, 'Visible','off');

cla(handles.axes_binary, 'reset')

set(handles.axes_binary, 'Visible','off');

cla(handles.axes_features, 'reset')

set(handles.axes_features, 'Visible','off');

set(handles.output, 'Pointer', 'arrow');


axis(handles.axes_write,[1 400 1 400]); % 设定图轴范围

RectHandle = rectangle(handles.axes_write,'Position',[80,66,240,268],'LineStyle','--','EdgeColor','#a9a9a9');

MouseDraw();

三、运行结果

在这里插入图片描述

完整代码或仿真咨询QQ1575304183

以上是关于手写数字识别基于支持向量机SVM实现手写数字识别matlab源码含GUI的主要内容,如果未能解决你的问题,请参考以下文章

手写数字识别基于matlab GUI SVM手写数字识别含Matlab源码 676期

Python,OpenCV基于支持向量机SVM的手写数字OCR

OpenCV-Python实战(番外篇)——利用 SVM 算法识别手写数字

OpenCV-Python实战(番外篇)——利用 SVM 算法识别手写数字

支持向量机

基于多层感知机的手写数字识别(Tensorflow实现)