-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathPlotData.m
289 lines (287 loc) · 13.2 KB
/
PlotData.m
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
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
%% 路径
tic
clear all % 清空工作区
clc % 清空命令行
real_path = 'E:\det\'; % 真实数据的路径\
calculate_rootpath = 'E:\result\'; % 预测结果的路径
% my_anchor_samplenum
CIOU_DIOU_NMS_5000_path = [calculate_rootpath, 'samplenum\CIOU+DIOU-NMS_5000\detect_results\'];
CIOU_DIOU_NMS_10000_path = [calculate_rootpath, 'samplenum\CIOU+DIOU-NMS_10000\detect_results\'];
CIOU_DIOU_NMS_20000_path = [calculate_rootpath, 'samplenum\CIOU+DIOU-NMS_20000\detect_results\'];
% my_anchor_strategy
CIOU_DIOU_NMS_20000_path = [calculate_rootpath, 'strategy\CIOU+DIOU-NMS_20000\detect_results\'];
CIOU_IOU_NMS_20000_path = [calculate_rootpath, 'strategy\CIOU+IOU-NMS_20000\detect_results\'];
IOU_NMS_20000_path = [calculate_rootpath, 'strategy\IOU-NMS_20000\detect_results\'];
Frcnn_path = 'E:\result\frcnn\';
%% 画出不同信噪比下三个检测指标的图
% 定义接收计算好的检测参数
num = 1000; % 每种db下文本文件的数量
CIOU_DIOU_NMS_5000 = []; % 5000张数据集[dB, Pd, Pf, Eavg]
CIOU_DIOU_NMS_10000 = []; % 10000张数据集[dB, Pd, Pf, Eavg]
CIOU_DIOU_NMS_20000 = []; % 20000张数据集[dB, Pd, Pf, Eavg]
Frcnn_20000 = []; % 用frcnn训练出的结果[dB, Pd, Pf, Eavg]
%% CIOU_DIOU_NMS_5000计算参数
for dB = 0:2:16 % 遍历每个SNR下的label文件夹
N3ME_5000 = []; % 计算完一个dB后刷新
for i = 1:num % 遍历每个SNR下的label文件夹下的一共num（1000）个labels文本文件，包括预测和实际
% 判断文件是否存在，存在就读取文件。yolov5如果一张图检测不到目标就不会输出txt文件。
if(exist([[CIOU_DIOU_NMS_5000_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']], 'file') == 2)
ML = PdPfEavg([[real_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']],...
[[CIOU_DIOU_NMS_5000_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']]);
% N3ME_5000 : [N, M, Mt, Mf, Eavg]
N3ME_5000 = [N3ME_5000; ML];
else % 不存在就记录下对应的实际信号的个数并开始下一个label文件
real_data = load([[real_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']]);
N = length(real_data(:,1)); % 一张图中实际信号的个数
N3ME_5000 = [N3ME_5000; N, 0, 0, 0, 0]; % 记录对应的实际信号的个数N并填充矩阵行数
disp(['Yolov5,数据集5000中', [num2str(dB),'dB下第', num2str(i), '张一个都没有检测出来']])
continue;
end
end
N3ME_5000_dB(:, :, dB/2 + 1) = N3ME_5000; % 保存所有db下的结果
N = sum(N3ME_5000(:, 1)); M = sum(N3ME_5000(:, 2));
Mt = sum(N3ME_5000(:, 3)); Mf = sum(N3ME_5000(:, 4));
Eavg = sum(N3ME_5000(:, 5));
% disp([num2str(dB),'dB下实际信号数量：', num2str(N)])
% disp([num2str(dB),'dB下检测出的信号数量：', num2str(M)])
% disp([num2str(dB),'dB下真实信号数量：', num2str(Mt)])
% disp([num2str(dB),'dB下虚假信号数量：', num2str(Mf)])
Pd = Mt/N; % 检测概率
Pf = Mf/M; % 虚警概率
Eavg = Eavg/(4 * Mt); % 信息参数平均误差
CIOU_DIOU_NMS_5000 = [CIOU_DIOU_NMS_5000; dB, Pd, Pf, Eavg];
end
%% CIOU_DIOU_NMS_10000计算参数
for dB = 0:2:16 % 遍历每个SNR下的label文件夹
N3ME_10000 = [];
for i = 1:num % 遍历每个SNR下的label文件夹下的一共num（1000）个labels文本文件，包括预测和实际
% 判断文件是否存在，存在就读取文件。yolov5如果一张图检测不到目标就不会输出txt文件。
if(exist([[CIOU_DIOU_NMS_10000_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']], 'file') == 2)
ML = PdPfEavg([[real_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']],...
[[CIOU_DIOU_NMS_10000_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']]);
% N3ME_10000 : [N, M, Mt, Mf, Eavg]
N3ME_10000 = [N3ME_10000; ML];
else % 不存在就记录下对应的实际信号的个数并开始下一个label文件
real_data = load([[real_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']]);
N = length(real_data(:,1)); % 一张图中实际信号的个数
N3ME_10000 = [N3ME_10000; N, 0, 0, 0, 0];
disp(['Yolov5,数据集10000中', [num2str(dB),'dB下第', num2str(i), '张一个都没有检测出来']])
continue;
end
end
N3ME_10000_dB(:, :, dB/2 + 1) = N3ME_10000; % 保存所有db下的结果
N = sum(N3ME_10000(:, 1)); M = sum(N3ME_10000(:, 2));
Mt = sum(N3ME_10000(:, 3)); Mf = sum(N3ME_10000(:, 4));
Eavg = sum(N3ME_10000(:, 5));
% disp([num2str(dB),'dB下实际信号数量：', num2str(N)])
% disp([num2str(dB),'dB下检测出的信号数量：', num2str(M)])
% disp([num2str(dB),'dB下真实信号数量：', num2str(Mt)])
% disp([num2str(dB),'dB下虚假信号数量：', num2str(Mf)])
Pd = Mt/N; % 检测概率
Pf = Mf/M; % 虚警概率
Eavg = Eavg/(4 * Mt); % 信息参数平均误差
CIOU_DIOU_NMS_10000 = [CIOU_DIOU_NMS_10000; dB, Pd, Pf, Eavg];
end
%% CIOU_DIOU_NMS_20000计算参数
for dB = 0:2:16 % 遍历每个SNR下的label文件夹
N3ME_20000 = [];
for i = 1:num % 遍历每个SNR下的label文件夹下的一共num（1000）个labels文本文件，包括预测和实际
% 判断文件是否存在，存在就读取文件。yolov5如果一张图检测不到目标就不会输出txt文件。
if(exist([[CIOU_DIOU_NMS_20000_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']], 'file') == 2)
ML = PdPfEavg([[real_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']],...
[[CIOU_DIOU_NMS_20000_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']]);
N3ME_20000 = [N3ME_20000; ML];
else % 不存在就记录下对应的实际信号的个数并开始下一个label文件
real_data = load([[real_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']]);
N = length(real_data(:,1)); % 一张图中实际信号的个数
N3ME_20000 = [N3ME_20000; N, 0, 0, 0, 0];
disp(['Yolov5,数据集20000中', [num2str(dB),'dB下第', num2str(i), '张一个都没有检测出来']])
continue;
end
end
N3ME_20000_dB(:, :, dB/2 + 1) = N3ME_20000; % 保存所有db下的结果
N = sum(N3ME_20000(:, 1)); M = sum(N3ME_20000(:, 2));
Mt = sum(N3ME_20000(:, 3)); Mf = sum(N3ME_20000(:, 4));
Eavg = sum(N3ME_20000(:, 5));
% disp([num2str(dB),'dB下实际信号数量：', num2str(N)])
% disp([num2str(dB),'dB下检测出的信号数量：', num2str(M)])
% disp([num2str(dB),'dB下真实信号数量：', num2str(Mt)])
% disp([num2str(dB),'dB下虚假信号数量：', num2str(Mf)])
Pd = Mt/N; % 检测概率
Pf = Mf/M; % 虚警概率
Eavg = Eavg/(4 * Mt); % 信息参数平均误差
CIOU_DIOU_NMS_20000 = [CIOU_DIOU_NMS_20000; dB, Pd, Pf, Eavg];
end
%% Frcnn计算参数
for dB = 0:2:16 % 遍历每个SNR下的label文件夹
N3ME_Frcnn = []; % 计算完一个dB后刷新
for i = 1:num % 遍历每个SNR下的label文件夹下的一共num（1000）个labels文本文件，包括预测和实际
% 判断文件是否存在，存在就读取文件。yolov5如果一张图检测不到目标就不会输出txt文件。
if(exist([[Frcnn_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']], 'file') == 2)
ML = PdPfEavg([[real_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']],...
[[Frcnn_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']]);
% N3ME_5000 : [N, M, Mt, Mf, Eavg]
N3ME_Frcnn = [N3ME_Frcnn; ML];
else % 不存在就记录下对应的实际信号的个数并开始下一个label文件
real_data = load([[real_path, num2str(dB), 'dB\labels\'],[num2str(i),'.txt']]);
N = length(real_data(:,1)); % 一张图中实际信号的个数
N3ME_Frcnn = [N3ME_Frcnn; N, 0, 0, 0, 0]; % 记录对应的实际信号的个数N并填充矩阵行数
disp(['Frcnn,数据集20000中', [num2str(dB),'dB下第', num2str(i), '张一个都没有检测出来']])
continue;
end
end
N3ME_Frcnn_dB(:, :, dB/2 + 1) = N3ME_Frcnn; % 保存所有db下的结果
N = sum(N3ME_Frcnn(:, 1)); M = sum(N3ME_Frcnn(:, 2));
Mt = sum(N3ME_Frcnn(:, 3)); Mf = sum(N3ME_Frcnn(:, 4));
Eavg = sum(N3ME_Frcnn(:, 5));
% disp([num2str(dB),'dB下实际信号数量：', num2str(N)])
% disp([num2str(dB),'dB下检测出的信号数量：', num2str(M)])
% disp([num2str(dB),'dB下真实信号数量：', num2str(Mt)])
% disp([num2str(dB),'dB下虚假信号数量：', num2str(Mf)])
Pd = Mt/N; % 检测概率
Pf = Mf/M; % 虚警概率
Eavg = Eavg/(4 * Mt); % 信息参数平均误差
Frcnn_20000 = [Frcnn_20000; dB, Pd, Pf, Eavg];
end
% CIOU_IOU_NMS_20000 = PdPfEavg(real_path, CIOU_IOU_NMS_20000_path);
% IOU_NMS_20000 = PdPfEavg(real_path, IOU_NMS_20000);
%% Plot
x_axis = 0:2:16;
% CIOU_DIOU_NMS_5000
CIOU_DIOU_NMS_5000_Pd = CIOU_DIOU_NMS_5000(:, 2);
CIOU_DIOU_NMS_5000_Pf = CIOU_DIOU_NMS_5000(:, 3);
CIOU_DIOU_NMS_5000_Eavg = CIOU_DIOU_NMS_5000(:, 4);
% CIOU_DIOU_NMS_10000
CIOU_DIOU_NMS_10000_Pd = CIOU_DIOU_NMS_10000(:, 2);
CIOU_DIOU_NMS_10000_Pf = CIOU_DIOU_NMS_10000(:, 3);
CIOU_DIOU_NMS_10000_Eavg = CIOU_DIOU_NMS_10000(:, 4);
% CIOU_DIOU_NMS_20000
CIOU_DIOU_NMS_20000_Pd = CIOU_DIOU_NMS_20000(:, 2);
CIOU_DIOU_NMS_20000_Pf = CIOU_DIOU_NMS_20000(:, 3);
CIOU_DIOU_NMS_20000_Eavg = CIOU_DIOU_NMS_20000(:, 4);
% Frcnn
Frcnn_20000_Pd = Frcnn_20000(:, 2);
Frcnn_20000_Pf = Frcnn_20000(:, 3);
Frcnn_20000_Eavg = Frcnn_20000(:, 4);
%% yolov5 不同训练数量的检测参数对比
% yolov5 Pd
figure(1) % 检测参数的图
subplot(131)
plot(x_axis, CIOU_DIOU_NMS_5000_Pd*100,'-o','MarkerIndices',1:length(x_axis));
hold on
plot(x_axis, CIOU_DIOU_NMS_10000_Pd*100, '-x','MarkerIndices',1:length(x_axis));
plot(x_axis, CIOU_DIOU_NMS_20000_Pd*100, '-*','MarkerIndices',1:length(x_axis));
set(gca,'XLim',[0 16]); % X轴的数据显示范围
set(gca, 'XTick', 0:2:16); % 设置要显示坐标刻
title('Pd');
xlabel('信噪比（dB）');
ylabel('检测概率（%）');
legend('训练时频图=5000','训练时频图=10000', '训练时频图=20000');
grid on % 加网格
% yolov5 Pf
subplot(132)
plot(x_axis, CIOU_DIOU_NMS_5000_Pf*100, '-o','MarkerIndices',1:length(x_axis));
hold on
plot(x_axis, CIOU_DIOU_NMS_10000_Pf*100, '-x','MarkerIndices',1:length(x_axis));
plot(x_axis, CIOU_DIOU_NMS_20000_Pf*100, '-*','MarkerIndices',1:length(x_axis));
set(gca,'XLim',[0 16]); % X轴的数据显示范围
set(gca, 'XTick', 0:2:16); % 设置要显示坐标刻
title('Pf');
xlabel('信噪比（dB）');
ylabel('虚警概率（%）');
legend('训练时频图=5000','训练时频图=10000', '训练时频图=20000');
grid on
% yolov5 Eavg
subplot(133)
plot(x_axis, CIOU_DIOU_NMS_5000_Eavg*100, '-o','MarkerIndices',1:length(x_axis));
hold on
plot(x_axis, CIOU_DIOU_NMS_10000_Eavg*100, '-x','MarkerIndices',1:length(x_axis));
plot(x_axis, CIOU_DIOU_NMS_20000_Eavg*100, '-*','MarkerIndices',1:length(x_axis));
set(gca,'XLim',[0 16]); % X轴的数据显示范围
set(gca, 'XTick', 0:2:16); % 设置要显示坐标刻
set(gca,'YLim',[0 8]); % Y轴的数据显示范围
title('Eavg');
xlabel('信噪比（dB）');
ylabel('参数估计平均误差（%）');
legend('训练时频图=5000','训练时频图=10000', '训练时频图=20000');
grid on
%% yolov5和Frcnn 检测参数对比
figure(2)
subplot(131) % Pd
plot(x_axis, CIOU_DIOU_NMS_20000_Pd*100, '-*','MarkerIndices',1:length(x_axis));
hold on
plot(x_axis, Frcnn_20000_Pd*100, '-x','MarkerIndices',1:length(x_axis));
set(gca,'XLim',[0 16]); % X轴的数据显示范围
set(gca, 'XTick', 0:2:16); % 设置要显示坐标刻
title('Pd');
xlabel('信噪比（dB）');
ylabel('检测概率（%）');
legend('YOLOV5', 'FRCNN');
grid on % 加网格
subplot(132) % Pf
plot(x_axis, CIOU_DIOU_NMS_20000_Pf*100, '-*','MarkerIndices',1:length(x_axis));
hold on
plot(x_axis, Frcnn_20000_Pf*100, '-x','MarkerIndices',1:length(x_axis));
set(gca,'XLim',[0 16]); % X轴的数据显示范围
set(gca, 'XTick', 0:2:16); % 设置要显示坐标刻
title('Pf');
xlabel('信噪比（dB）');
ylabel('虚警概率（%）');
legend('YOLOV5', 'FRCNN');
grid on
subplot(133) % Eavg
plot(x_axis, CIOU_DIOU_NMS_20000_Eavg*100, '-*','MarkerIndices',1:length(x_axis));
hold on
plot(x_axis, Frcnn_20000_Eavg*100, '-x','MarkerIndices',1:length(x_axis));
set(gca,'XLim',[0 16]); % X轴的数据显示范围
set(gca, 'XTick', 0:2:16); % 设置要显示坐标刻
set(gca,'YLim',[0 8]); % Y轴的数据显示范围
title('Eavg');
xlabel('信噪比（dB）');
ylabel('参数估计平均误差（%）');
legend('YOLOV5', 'FRCNN');
grid on
%% 直方分析图
figure(3)
N3ME_5000_dB_Mt = [];
N3ME_10000_dB_Mt = [];
N3ME_20000_dB_Mt = [];
% 各个数据集不同db下真实信号的个数
bar_x = 0:2:16;
% 训练集5000,10000,20000，0-16db下的真实信号
% [N, M, Mt, Mf, Eavg]
for db = 0:2:16
N3ME_5000_dB_Mt = [N3ME_5000_dB_Mt, sum(N3ME_5000_dB(:,4,db/2 + 1))];
N3ME_10000_dB_Mt = [N3ME_10000_dB_Mt, sum(N3ME_10000_dB(:,4,db/2 + 1))];
N3ME_20000_dB_Mt = [N3ME_20000_dB_Mt, sum(N3ME_20000_dB(:,4,db/2 + 1))];
end
bar_y = [N3ME_5000_dB_Mt(1), N3ME_10000_dB_Mt(1), N3ME_20000_dB_Mt(1); % 0dB
N3ME_5000_dB_Mt(2), N3ME_10000_dB_Mt(2), N3ME_20000_dB_Mt(2); % 2dB
N3ME_5000_dB_Mt(3), N3ME_10000_dB_Mt(3), N3ME_20000_dB_Mt(3); % 4dB
N3ME_5000_dB_Mt(4), N3ME_10000_dB_Mt(4), N3ME_20000_dB_Mt(4); % 6dB
N3ME_5000_dB_Mt(5), N3ME_10000_dB_Mt(5), N3ME_20000_dB_Mt(5); % 8dB
N3ME_5000_dB_Mt(6), N3ME_10000_dB_Mt(6), N3ME_20000_dB_Mt(6); % 10dB
N3ME_5000_dB_Mt(7), N3ME_10000_dB_Mt(7), N3ME_20000_dB_Mt(7); % 12dB
N3ME_5000_dB_Mt(8), N3ME_10000_dB_Mt(8), N3ME_20000_dB_Mt(8); % 14dB
N3ME_5000_dB_Mt(9), N3ME_10000_dB_Mt(9), N3ME_20000_dB_Mt(9);]; % 16dB
b = bar(bar_x, bar_y, 'FaceColor', 'flat');
for k = 1:size(bar_y,2)
b(k).CData = k;
end
xtips1 = b(1).XEndPoints;
ytips1 = b(1).YEndPoints;
labels1 = string(b(1).YData);
text(xtips1,ytips1,labels1,'HorizontalAlignment','center',...
'VerticalAlignment','bottom')
xtips2 = b(2).XEndPoints;
ytips2 = b(2).YEndPoints;
labels2 = string(b(2).YData);
text(xtips2,ytips2,labels2,'HorizontalAlignment','center',...
'VerticalAlignment','bottom')
xtips3 = b(3).XEndPoints;
ytips3 = b(3).YEndPoints;
labels3 = string(b(3).YData);
text(xtips3,ytips3,labels3,'HorizontalAlignment','center',...
'VerticalAlignment','bottom')
legend('5000','10000', '20000');
toc