制作VOC数据集

• 摄像头采集样张

使用opencv，外接摄像头，按一定帧率采集图像，代码如下：

import cv2 as cv
cap = cv.VideoCapture(0)

fourcc = cv.VideoWriter_fourcc('X', 'V', 'I', 'D')
out = cv.VideoWriter(r'C:\Users\chen\Desktop\PyTorch练习\output.avi', fourcc, 20.0, (640,480))
i = 0
while cap.isOpened():
    i = i+1
    ret, frame = cap.read()
    if not ret:
        break
    frame = cv.flip(frame, 1)
    if i%10 == 0:
        j = i//10
        cv.imwrite(str(j) + '.jpg', frame)
    #cv.line(frame, (0,0),(639,479),(255,0,0),15)
    out.write(frame)
    cv.imshow('frame', frame)
    if cv.waitKey(1) == ord('q'):
        break
    
cap.release()
out.release()
cv.destroyAllWindows()

 

• 标注

工具Windows10+Python3.7+LableImg

1.下载LableImg

链接：https://pan.baidu.com/s/1u-S42NUrDeyAz53IE7E_3g 密码：hxsz

2.安装Python

3.配置环境

打开cmd，依次输入

pip install PyQt5

　　

pip install pyqt5-tools

　　

pip install lxml

4.标注

双击软件，开始标注（在/data/文件中编辑自己的类别）

• 生成Main文件中的txt文件

在数据集根目录下，我的是Mydataset文件夹，按住shift+鼠标右键，选择powershell，键入ipython

输入如下代码，即可在Main文件中生成4个txt文件。一共是2-19，共18张图，train9个（12x0.8=9.6=9），test6个，val3个，trainval12个（18x0.7=12.6=12）

import os  
import random  

trainval_percent = 0.7  
train_percent = 0.8
xmlfilepath = 'Annotations/'
txtsavepath = 'ImageSets/Main'  
total_xml = os.listdir(xmlfilepath)  

num = len(total_xml)  
list = range(num)  
tv = int(num*trainval_percent)  
tr = int(tv*train_percent)  
trainval = random.sample(list,tv)  
train = random.sample(trainval,tr)  

ftrainval = open(txtsavepath+'/trainval.txt', 'w')  
ftest = open(txtsavepath+'/test.txt', 'w')  
ftrain = open(txtsavepath+'/train.txt', 'w')  
fval = open(txtsavepath+'/val.txt', 'w')  

for i in list:  
    name = total_xml[i][:-4]+'\n'  
    if i in trainval:  
        ftrainval.write(name)  
        if i in train:  
            ftrain.write(name)  
        else:  
            fval.write(name)  
    else:  
        ftest.write(name)  

ftrainval.close()  
ftrain.close()  
fval.close()  
ftest .close()
print('Well Done！！！')