【行为识别】基于matlab轨迹法行为识别【含Matlab源码 375期】

一、获取代码方式

获取代码方式1：
通过订阅紫极神光博客付费专栏，凭支付凭证，私信博主，可获得此代码。

获取代码方式2：
完整代码已上传我的资源：【行为识别】基于matlab轨迹法行为识别【含Matlab源码 375期】

备注：
订阅紫极神光博客付费专栏，可免费获得1份代码（有效期为订阅日起，三天内有效）；

二、部分源代码

%%%%%%%
%%% Main training/testing file
close all
clear all;
clc;

IraniDatasetPath = 'C:\Users\lenovo\Desktop\17351584MACH3D-Rodriguez\MACH3D\actions\bend';
trainAction = 'bend';
threshold = 0.73; % bend: 0.725; pjump: 0.76; runleft: 0.73; jack: 0.77; wave1: 0.78
actions = {'bend', 'jack', 'jump', 'pjump', 'run', 'runleft', 'runright', 'side', 'skip', 'walk', 'wave1', 'wave2'};   %  'runleft', 'runright'
actors = {'daria', 'denis', 'eli', 'ido', 'ira', 'lena', 'lyova', 'moshe', 'shahar'};
trainActorID = [1 2 3 4 6 9]; % [1 2 3 4 6 9];   % [1 2 3 4 6 9];    % i.e. Training will be done on: daria, denis, eli, ido, lena, shahar 
testActorID = [5]; % [1, 7, 8];    %s[5 7 8]               % i.e. Testing of the trained 3D MACH filter will be done on all the actions done by: ira, lyova, and moshe

% programRootPath =  'D:\Mikel\mFiles\Image Processing\MACH3D\';

numActions = length(actions);
numTestActors = length(testActorID);
numTrainActors = length(trainActorID);
volumes = cell(1, numTrainActors);

for v = 1 : numTrainActors
    inFile = sprintf('%s_%0.3d.avi', trainAction, trainActorID(v));
    % inFile = [programRootPath, inFile];
    ifp = aviinfo(inFile);
    volume = zeros(ifp.Height, ifp.Width, ifp.NumFrames, 'uint8');
    for f = 1 : ifp.NumFrames
        frame = aviread(inFile, f);
        rgbImg = frame.cdata;
        grayImg = rgb2gray(rgbImg);
        edgeImg = sobel(grayImg);
        volume(:,:,f) = edgeImg;
    end
    volumes{v} = volume;
end
mach3d = train_otmach3(volumes);
mach_file = sprintf('%s_mach.mat', trainAction);
save (mach_file, 'mach3d')


% Save 3D MACH as a short movie clip

machMovieFile = sprintf('%s_mach.avi', trainAction);
mov = avifile(machMovieFile, 'COMPRESSION', 'None', 'FPS', ifp.FramesPerSecond, 'QUALITY', 100);  % "Indeo5" is better and offer more compression than "Cinepak"
figure(1);
for f = 1 : ifp.NumFrames 
    rgbMACH = cat(3, mach3d(:,:,f), mach3d(:,:,f), mach3d(:,:,f));
    m = im2frame(rgbMACH);
    mov = addframe(mov, m);
    imshow(rgbMACH);
    pause(0.040);
end
disp('3D MACH Filter as a Movie');
disp('=========================');
mov = close(mov)


% Test 3-D MACH filter
% =======================
for a = 1 : numActions
    for v = 1 : numTestActors

        inFileTest = [IraniDatasetPath, sprintf('%s\\%s_%s.avi', actions{a}, actors{testActorID(v)}, actions{a})];
       

        iftp = aviinfo(inFileTest);
        volume = zeros(iftp.Height, iftp.Width, iftp.NumFrames, 'uint8');
        figure(2);
        for f = 1 : iftp.NumFrames
            frame = aviread(inFileTest, f);
            rgbImg = frame.cdata;
            grayImg = rgb2gray(rgbImg);
            edgeImg = sobel(grayImg);
            volume(:,:,f) = edgeImg;
    
        end
        tic
        c = fftnormxcorr3(volume, mach3d);
        disp('');
        disp(sprintf('Test Clip: %s_%s.avi', actors{testActorID(v)}, actions{a}));
        disp('============================')
        [cmax, i, j, k] = max3(c)
        toc

        lineWidth = 4;
        color = [255 255 0];


        outFile = sprintf('Result_TrainedOn_%s_TestedOn_%s_%s.avi', trainAction, actions{a}, actors{testActorID(v)}); 

        mov = avifile(outFile, 'COMPRESSION', 'Indeo5', 'FPS', iftp.FramesPerSecond, 'QUALITY', 100);  % "Indeo5" is better and offer more compression than "Cinepak"

        figure(4)
        for f = 1 : iftp.NumFrames - ifp.NumFrames + 1
            [cmax2, i2, j2] = max2(c(:,:,f));
            frame = aviread(inFileTest, f);
            rgbImg = frame.cdata;
            if cmax2 > threshold
                rgbImg = putColorRectangle(rgbImg, i2, j2, i2+ifp.Height-1, j2+ifp.Width-1, lineWidth, color);
                rgbImg = putColorCrossHair(rgbImg, 'rgb', round(i2+ifp.Height/2), round(j2+ifp.Width/2), 25, 25, 3, color);
            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
  
96
  
97
  
98
  
99
  
100
  
101
 

三、运行结果

四、matlab版本及参考文献

1 matlab版本
2014a

2 参考文献
[1] 蔡利梅.MATLAB图像处理——理论、算法与实例分析[M].清华大学出版社，2020.
[2]杨丹,赵海滨,龙哲.MATLAB图像处理实例详解[M].清华大学出版社，2013.
[3]周品.MATLAB图像处理与图形用户界面设计[M].清华大学出版社，2013.
[4]刘成龙.精通MATLAB图像处理[M].清华大学出版社，2015.

文章来源: qq912100926.blog.csdn.net，作者：海神之光，版权归原作者所有，如需转载，请联系作者。

原文链接：qq912100926.blog.csdn.net/article/details/113995288