《玩儿起来吧》MATLAB 实时图像处理系列（一二三四）_matlab如何获取实时的图像

大家学了图像处理之后，是不是就想学以致用。而要用于生活中，最好的方式就是和摄像头搭配，实现实时图像处理。今天就给大家分享MATLAB实时图像处理的基础方法，希望能帮你快速入门。

by 今天不飞了

---

视频教学地址

《玩儿起来吧》MATLAB实时图像处理（一）如何获取实时图像
《玩儿起来吧》MATLAB实时图像处理（二）如何实时处理图像
《玩儿起来吧》MATLAB实时图像处理（三）最简单的人脸检测与追踪
《玩儿起来吧》MATLAB实时图像处理（四）目标追踪

---

一、如何获取实时图像

1. 有摄像头的小伙伴
   这篇博客有基本的摄像头配置方法 《MATLAB摄像头实时图像处理》

%% 配置摄像头
% -----------------------------------------
% 查看适配器
disp(imaqhwinfo)
% 查看设备及其支持的格式
info = imaqhwinfo('winvideo');
disp(info.DeviceInfo.SupportedFormats)
% --------你要是知道就不要运行这两句-------

%% 生成对象并同步画面
obj = videoinput('winvideo',1,'MJPG_640x480');
h = preview(obj);

%% 实时显示（处理）
figure('Position',[800,400,1000,500])
while ishandle(h)
    tic
    % 获取影像
    frame = getsnapshot(obj);  % 获取帧
    % 一顿疯狂处理
    img = rgb2gray(frame);
    % 显示
    subplot(121),imshow(frame)
    subplot(122),imshow(img)
    drawnow
    t = toc;
    disp(round(1/t))
end
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

2. 没有摄像头的小伙伴
   看这里【MATLAB】没有摄像头怎么做实时图像处理

---

二、如何实时处理图像

答案就是，利用while语句，对获取到的影像“一顿疯狂处理”。
请自行加入自己的算法。

%% 实时显示（处理）
% 生成对象并开启摄像机
obj = videoinput('winvideo',1,'MJPG_640x480'); 
h1 = preview(obj);
h2 = figure('Position',[100,50,1680,800]);

% 开始
while ishandle(h1)&&ishandle(h2)
    tic
    % 获取影像
    frame = getsnapshot(obj);  % 获取帧
    frame = im2double(frame(:,1+80:end-80,:));
    
    % 一顿疯狂处理
    % --------------------------------
    img1 = DoSomethingCrazy1(frame);
    img2 = DoSomethingCrazy2(frame);
    img3 = frame.*img2*2;
    % --------------------------------    
    img = cat(2,img1,cat(3,img2,img2,img2),img3);
    % 保存
    imshow(img)
    drawnow

    t = toc;
    disp(round(1/t))
end


function ed = DoSomethingCrazy1(frame)
% 归一化彩色空间
gray = rgb2gray(frame);
mask = double(gray>0.05);
imsum = sqrt(sum(frame.^2,3));
ed = frame./imsum.*mask;
end

function ed = DoSomethingCrazy2(frame)
% 灰度边缘
core1 = [1,1,1;0,0,0;-1,-1,-1;];
core2 = [1,1,0;1,0,-1;0,-1,-1;];
frame = rgb2gray(frame);
im1 = imfilter(frame,core1);
im2 = imfilter(frame,core1');
im3 = imfilter(frame,core2);
im4 = imfilter(frame,core2');
ed = max(abs(cat(3,im1,im2,im3,im4)),[],3);
end
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48

---

三、简单的人脸识别

核心函数vision.CascadeObjectDetector，使用方法演示代码如下

function findme()

% 创建GUI
Fig = figure('Position',[100,150,980,500]);

Pnl1 = uipanel(Fig,'Position',[0.05,0.17,0.9,0.8]);
Pnl2 = uipanel(Fig,'Position',[0.05,0.05,0.9,0.1]);

Axes1 = axes(Pnl1,'Position',[0,0,1/2,1]);
Axes2 = axes(Pnl1,'Position',[1/2,0,1/2,1]);

Bt = uicontrol(Pnl2,'style','togglebutton','String','人脸检测','Fontsize',16,...
    'Units','normalized','Position',[2/5,0,1/5,1],'Callback',@FaceDetection);
drawnow

% 开启摄像头
Hcamera = [];
Hobj = [];
if isempty(Hcamera)
    Hobj = videoinput('winvideo',1,'MJPG_640x480');
    Hcamera = preview(Hobj);
    frame = getsnapshot(Hobj);  % 获取帧
    [rows,cols,~] = size(frame);
end

% 人脸识别
flag = 0;
faceDetector = vision.CascadeObjectDetector();
gap = 2;

% 画
[Face,rp,cp] = drawface();

% 开始
while 1
    if ishandle(Hcamera)
        % 获取影像
        frame = getsnapshot(Hobj);  % 获取帧
        frame = im2double(frame);
        % 显示
        imshow(frame,'Parent',Axes1)
        % 检测
        if flag
            bboxx = face(frame);
            if ~isempty(bboxx)
                Px = bboxx(1,1);
                Py = -bboxx(1,2);
                Face.XData = Px;
                Face.YData = Py;
            end
            hold(Axes1,'on')
            for i = 1:size(bboxx,1)
                bbox = bboxx(i,:);
                rc = bbox+[-bbox(3)/4,-bbox(4)/4,bbox(3)/2,bbox(4)/2];
                rectangle('Position',rc,'Curvature',0,...
                    'LineWidth',2,'LineStyle','--',...
                    'EdgeColor','y','Parent',Axes1)
            end
            hold(Axes1,'off')
        end
        drawnow
    else
        break
    end
end

    function FaceDetection(~,~)
        flag = get(Bt,'Value');
    end

    function bboxx = face(frame)
        frame = frame(1:gap:end,1:gap:end,:);
        bboxx = step(faceDetector, frame);
        bboxx = bboxx*gap;
    end

    function [h,rp,cp] = drawface()        
        rp = cols/2;
        cp = -rows/2;
        % 绘制
        h = plot(rp,cp,'ro','MarkerSize', 35);
        hold off
        axis equal
        axis([0,cols,-rows,0,0,1])
        axis off
        view([0,0,1])
        set(gca,'looseInset',[0 0 0 0])
    end

end
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90

想加入b站表情包，看这篇把B站表情包植入MATLAB

---

四、简单的目标追踪

核心函数vision.PointTracker('MaxBidirectionalError',1)，[point,validity] = tracker(frame)，使用方法演示代码如下

function findme()

% 创建GUI
Fig = figure('Position',[500,450,980,500]);

Pnl1 = uipanel(Fig,'Position',[0.05,0.17,0.9,0.8]);
Pnl2 = uipanel(Fig,'Position',[0.05,0.05,0.9,0.1]);

Axes1 = axes(Pnl1,'Position',[0,0,1/2,1]);
Axes2 = axes(Pnl1,'Position',[1/2,0,1/2,1]);

Bt = uicontrol(Pnl2,'style','togglebutton','String','锁定目标','Fontsize',16,...
    'Units','normalized','Position',[2/5,0,1/5,1],'Callback',@LockTarget);
drawnow

% 开启摄像头
Hcamera = [];
Hobj = [];
if isempty(Hcamera)
    Hobj = videoinput('winvideo',1,'MJPG_640x480');
    Hcamera = preview(Hobj);
    frame = getsnapshot(Hobj);  % 获取帧
end

% 人脸识别
flag = 0;
tracker = vision.PointTracker('MaxBidirectionalError',1);
gap = 1;
objectRegion = [0,0,0,0];
objFrame = [];
points = [];


% 开始
while 1
    if ishandle(Hcamera)
        % 获取影像
        frame = getsnapshot(Hobj);  % 获取帧
        frame = im2double(frame);
        if flag
            % 检测
            point = face(frame);
            objimg = insertMarker(frame,point,'+','Color','green');
            imshow(objimg,'Parent',Axes1)
        else
            % 普通显示
            imshow(frame,'Parent',Axes1)
        end
        drawnow
    else
        break
    end
end

    function LockTarget(~,~)
        % 降采样
        frame = frame(1:gap:end,1:gap:end,:);
        objFrame = frame;
        % 确定目标位置
        [x,y,~] = ginput(2);
        objectRegion = [min(x),min(y),max(x)-min(x),max(y)-min(y)]/gap;
        % 检测特征点
        points = detectMinEigenFeatures(rgb2gray(frame),'ROI',objectRegion);
        pointImage = insertMarker(frame, points.Location,'+','Color','green');
        imshow(pointImage,'Parent',Axes2)
        % 初始化跟踪器
        release(tracker)
        initialize(tracker, points.Location, frame);
        flag = 1;
    end

    function point = face(frame)
        % 降采样
        frame = frame(1:gap:end,1:gap:end,:);
        % 追踪
        [point,validity] = tracker(frame);
        point = point(validity,:);
        % 更新目标
        if size(point,1)<20
            % 方法一
%             x = point(:,1);
%             y = point(:,2);
%             objectRegion = [min(x),min(y),max(x)-min(x),max(y)-min(y)];
%             points = detectMinEigenFeatures(rgb2gray(frame),'ROI',objectRegion);
            % 方法二
            release(tracker)
            initialize(tracker, points.Location, objFrame);
            [point,validity] = tracker(frame);
            point = point(validity,:);
        end
        % 反算坐标
        point = point*gap;
    end

end
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95

其他

1. 不要直接照搬代码，你的摄像头型号参数不一定跟我一样，请根据实际情况修改。具体方法见以下视频或博客
   《玩儿起来吧》MATLAB实时图像处理（一）如何获取实时图像
   《MATLAB摄像头实时图像处理》
2. 暂时没想到