海思项目总结——python中OS

在制作训练集过程中，经常能遇到各种问题。解决这些问题，通过python对系统文件的操作至关重要。

1.菜鸟编程OS操作

2.

https://www.jb51.net/article/138812.htm

3.images 文件夹中删除labels文件夹中没有的相关项目

import os
import re
import sys
def contnameall():
	fileList = os.listdir(r"D:\Documents\Desktop\cat_dog_dataset\training_dataset\training_data\labels")		#待修改文件夹		#输出文件夹中包含的文件
	for num in range(1,2001,1):
		if 'cat_'+str(num)+'.txt' not in fileList:
			print('cat_'+str(num))
			try:
				os.remove('D:/Documents/Desktop/cat_dog_dataset/training_dataset/training_data/images/'+'cat_'+str(num)+'.jpg')
			except FileNotFoundError:
				continue
			
	for num in range(1,2001,1):
		if 'dog_'+str(num)+'.txt'not in fileList:
			print('dog_'+str(num))
			try:
				os.remove('D:/Documents/Desktop/cat_dog_dataset/training_dataset/training_data/images/'+'dog_'+str(num)+'.jpg')
			except FileNotFoundError:
				continue                        
			
contnameall()

4.修改文件名

import os
import re
import sys
def renameall():
    fileList = os.listdir(r"D:\Documents\Desktop\cat_dog_data\environment")        #待修改文件夹
    print("修改前："+str(fileList))        #输出文件夹中包含的文件
    currentpath = os.getcwd()        #得到进程当前工作目录
    os.chdir(r"D:\Documents\Desktop\cat_dog_data\environment")        #将当前工作目录修改为待修改文件夹的位置
    num=1        #名称变量
    for fileName in fileList:        #遍历文件夹中所有文件
        pat=".+\.(jpg|png|gif)"        #匹配文件名正则表达式
        pattern = re.findall(pat,fileName)        #进行匹配
        os.rename(fileName,('en_'+str(num)+'.'+pattern[0]))        #文件重新命名
        num = num+1        #改变编号，继续下一项
    print("---------------------------------------------------")
    os.chdir(currentpath)        #改回程序运行前的工作目录
    sys.stdin.flush()        #刷新
    print("修改后："+str(os.listdir(r"D:\Documents\Desktop\cat_dog_data\environment")))        #输出修改后文件夹中包含的文件
renameall()

5.json文件转txt文件

import os
import json
import numpy as np
import math
 
dir_json = 'D:/Documents/Desktop/cat_dog_dataset/validation_dataset/validation_data/json/' #json存储的文件目录
dir_txt = 'D:/Documents/Desktop/cat_dog_dataset/validation_dataset/validation_data/labels/'    #txt存储目录
if not os.path.exists(dir_txt):
    os.makedirs(dir_txt)
list_json = os.listdir(dir_json)

  
def json2txt(path_json,path_txt):    #可修改生成格式
    with open(path_json,'r') as path_json:
        jsonx=json.load(path_json)
        with open(path_txt,'w+') as ftxt:
            imageHeight = jsonx['imageHeight']
            imageWidth = jsonx['imageWidth']
            for shape in jsonx['shapes']:
                label = str(shape['label'])+' '
                xy=np.array(shape['points'])
                w = abs(xy[1][0]-xy[0][0])
                h = abs(xy[1][1]-xy[0][1])
                center_x = (xy[1][0]+xy[0][0])/2
                center_y = (xy[1][1]+xy[0][1])/2
                w = w/imageWidth
                h = h/imageHeight
                center_x = center_x/imageWidth
                center_y = center_y/imageHeight
                strxy = '\n'
                strxy=strxy+str(center_x)+'\n'+str(center_y)+'\n'+str(w)+'\n'+str(h)
                label += strxy
                ftxt.writelines(label + "\n")
                
'''               for m,n in xy:
                    m=int(m)
                    n=int(n)
                    strxy+=str(m)+' '+str(n)+' '
'''

 
 
for cnt,json_name in enumerate(list_json):
    print('cnt=%d,name=%s'%(cnt+1,json_name))
    path_json = dir_json + json_name
    path_txt = dir_txt + json_name.replace('.json','.txt')
    json2txt(path_json,path_txt)


#修改代码能够支持dark训练，归一化结果

6.将图像的P模式转化为RGB (labelme只能操作RGB 模式)

import os
from PIL import Image
 
# 将所有图像resize成224*224，并保存在目录'result/'下
def resize_image(image_name):
    # 获取输入文件夹中的所有文件
    files = os.listdir(image_name+'images/')
 
    output_dir = image_name+'images/'
    # 判断输出文件夹是否存在，不存在则创建
#    if not os.path.exists(output_dir):
#        os.makedirs(output_dir)
 
    for file in files:
        img = Image.open(image_name +'images/' + file)
        if img.mode == "P" or img.mode == 'RGBA':
            img = img.convert('RGB')
#        img = img.resize((224, 224), Image.ANTIALIAS)
        img.save(os.path.join(output_dir, file))
 
 
if __name__ == '__main__':
    resize_image('D:/Documents/Desktop/cat_dog_dataset/validation_dataset/validation_data/')

7.通过二进制形式读取文件

f = open('D:\Documents\Desktop\darknet-master\dog_cat_training\list.txt',"rb+")
data = f.read() #这样data是一个b开头的ASCII数字。
f.close()
print(data) #将二进制数据转化为10进制数据。

8.将image文件名读取到txt文件中

import os
def ListFilesToTxt(dir, file, wildcard, recursion):
    exts = wildcard.split(" ")
    files = os.listdir(dir)
    for name in files:
        fullname = os.path.join(dir, name)
        if (os.path.isdir(fullname) & recursion):
            ListFilesToTxt(fullname, file, wildcard, recursion)
        else:
            for ext in exts:
                if (name.endswith(ext)):
                    file.write("/home/hisi/data/dog_and_cat/image/"+name + "\n")
                    break

def ReadName():
    dir = "D:/Documents/Desktop/dog_and_cat/image/"  # 读取文件路径
    outfile = "D:/Documents/Desktop/dog_and_cat/list/list.txt"  # 将文件名写入Image.txt
    wildcard = ".jpg" # 读取kml文件
    #   wildcard = ".jpg .txt .exe .dll .lib"等等      #要读取的文件类型；
    file = open(outfile, "w")
    if not file:
        print("cannot open the file %s for writing" % outfile)
    ListFilesToTxt(dir, file, wildcard, 1)
    file.close()

ReadName()