%这里只是主函数
1 %{
2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3 % 可调参数
4
5 test_path='C:\Users\cong\Desktop\研一实战\项目\图像中时间数字识别\OCR\one\3.jpg';
6 neighbour_pixels_affect=3;
7 target_digit=2;
8 % forestTrain()参数设置
9 % .M - [1] number of trees to train
10 % .H - [max(hs)] number of classes
11 % .N1 - [5*N/M] number of data points for training each tree
12 % .F1 - [sqrt(F)] number features to sample for each node split
13 % .split - ['gini'] options include 'gini', 'entropy' and 'twoing'
14 % .minCount - [1] minimum number of data points to allow split
15 % .minChild - [1] minimum number of data points allowed at child nodes
16 % .maxDepth - [64] maximum depth of tree
17 % .dWts - [] weights used for sampling and weighing each data point
18 % .fWts - [] weights used for sampling features
19 % .discretize - [] optional function mapping structured to class labels
20 % format: [hsClass,hBest] = discretize(hsStructured,H);
21 varargin.M=1000;
22 %varargin.H=10;
23
24 % forestApply()的输入设置
25 % data - [NxF] N length F feature vectors
26 % forest - learned forest classification model
27 % maxDepth - [] maximum depth of tree
28 % minCount - [] minimum number of data points to allow split
29 % best - [0] if true use single best prediction per tree
30
31 % forestApply()输出结果及对比的阀值
32 % hs - [Nx1] predicted output labels
33 % ps - [NxH] predicted output label probabilities
34 ps_val_more_than0_3=0.2;
35
36 %滑窗检测,窗口尺度，步长
37 win_h=20;
38 win_w=20;
39 step=1;
40 disp('参数配置成功...');
41 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
42 disp('正在读入图片及特征提取...');
43 %读入图片及特征提取
44 data=[];
45 label=[];
46 temp_r1=0;
47 temp_c1=0;
48
49 for i_digit=0:9
50 % if(i_digit==target_digit) %%%%%%%%%%%%%%%%%%%%%%
51 % this_image_label=1;
52 % end
53 %数字转字符
54 str=num2str(i_digit); %%数据是不是不平衡
55 path_temp=strcat('C:\Users\cong\Desktop\研一实战\项目\图像中时间数字识别\trainingSample\num',str,'\');
56 file=dir(path_temp);
57 for i=3:length(file)
58 path= strcat(path_temp,file(i).name);
59
60 %%%%%%%%%%%%%%%%%%%%%%%%%%
61 % 加载图片
62 %%%%%%%%%%%%%%%%%%%%%%%%%%
63 I=imread(path);
64 %I=imread('E:/WeChat.jpg');
65 %%%%%%%%%%%%%%%%%%%%%%%%%%
66 % 提取channel features
67 %%%%%%%%%%%%%%%%%%%%%%%%%%
68 [all_channel_difference_features,temp_r1,temp_c1]=extract_features(I,neighbour_pixels_affect,1);
69 data=[data,all_channel_difference_features];
70 label=[label;i_digit+1];
71 if(rem(i,100)==0)
72 disp('...');
73 end
74 end % for i=3:length(file)
75 disp('数字')
76 i_digit
77 disp('的特征提取完毕...');
78 end % for i_digit=0:9
79 disp('读入图片及特征提取完毕...');
80 %%%%%%%%%%%%%%%%%%%%%%%%%%
81 % 扔进分类器中，训练
82 %%%%%%%%%%%%%%%%%%%%%%%%%%
83 data=data';
84 disp('正在训练，请稍等...');
85 forest = forestTrain( data, label, varargin );
86 disp('训练完毕...');
87 %}
88 %%%%%%%%%%%%%%%%%%%%%%%%%%
89 % 检测，测试
90 test_label=[];
91 test_label_p=[];
92 win_h=40;
93 win_w=30;
94 windSize = [30,40];
95 step=1;
96 ps_val_more_than0_3=0.07;
97
98 disp('正在检测...');
99 test_image=imread(test_path);
100 %滑窗检测,窗口尺度，步长
101 [test_r,test_c,test_z]=size(test_image);
102 figure;
103 imshow(test_image);
104 hold on
105
106 for j_test=1:step:(test_c-win_w+1)
107 for i_test=1:step:(test_r-win_h+1)
108 %model
109
110 model=test_image(i_test:i_test+win_h-1,j_test:j_test+win_w-1,:);
111 %resize
112 test_image_rs=imresize(model,[temp_r1 temp_c1]);
113 test_data=extract_features(test_image_rs,neighbour_pixels_affect,0);
114 test_data=test_data';
115 test_data=single(test_data);
116
117 [hs,ps] = forestApply( test_data, forest,0,0,1);%尺度问题
118 test_label=[test_label,hs];
119 test_label_p=[test_label_p,ps(hs)];
120 if(ps>ps_val_more_than0_3)
121 %画框
122 %%%%i_test
123 %j_test
124
125 rectangle('Position',[j_test,i_test,20,20],'LineWidth',4,'EdgeColor','r');
126 %pointAll = [i_test,j_test];
127 %[state,results]=draw_rect(test_image,pointAll,windSize);
128 hold on
129 end
130
131 end
132 end
133 disp('检测完毕！恭喜恭喜！')
134 %%%%%%%%%%%%%%%%%%%%%%%%%%
