深度学习中常用的工具脚本

1.创建txt文件的路径名

import os;
import shutil;
 
def listname(path,idtxtpath):
    filelist = os.listdir(path);# 该文件夹下所有的文件（包括文件夹）
    filelist.sort()
    f = open(idtxtpath, 'w');
    for files in filelist:  # 遍历所有文件
        Olddir = os.path.join(path, files);  # 原来的文件路径
        if os.path.isdir(Olddir):  # 如果是文件夹则跳过
            continue;
        f.write(path)
        f.write("/");
        f.write(files);
        f.write('\n');
    f.close();
 
savepath = os.getcwd()
imgidtxttrainpath = savepath+"/trainImageId.txt"
imgidtxtvalpath = savepath + "/validateImageId.txt"
listname(savepath + "/train/images",imgidtxttrainpath)
listname(savepath + "/val/images",imgidtxtvalpath)
print "trainImageId.txt && validateImageId.txt have been created!"

 

2.labelimg标签做好后，有些图片有标签，有些没有，这时候如果全部放到yolo中训练的话，某些图片找不到对应的txt，则会出错，该脚本删除那些没有标签的图片

import shutil;

def delete(path):
    savepath=os.getcwd()+"/saveimages";
    isExists=os.path.exists(savepath);
    if not isExists:
        os.makedirs(savepath,493);
    filelist=os.listdir(path);#该文件夹下所有的文件
    filelist.sort()
    for files in filelist:
        pos=-1;
        if "txt" in files:
            print files;
            filearray=files.split("-");
            number=filearray[1].split(".");
            picname=path+"/core-"+number[0]+".jpeg";
            shutil.move(picname,savepath);

savepath=os.getcwd();
delete(savepath+"/2");

 

3.xml转txt

注意yolo需要的格式为

# -*- coding: utf-8 -*-
import os
import sys
import xml.etree.ElementTree as ET
import glob

def convert(size,box):
    dw = 1./(size[0])
    dh = 1./(size[1])
    x = (box[0] + box[1])/2.0
    y = (box[2] + box[3])/2.0
    w = box[1] - box[0]
    h = box[3] - box[2]
    x = x*dw
    w = w*dw
    y = y*dh
    h = h*dh
    return (x, y, w, h)

def xml_to_txt(indir,outdir):

    os.chdir(indir)
    annotations = os.listdir('.')
    annotations = glob.glob(r"*.xml")

    for i, file in enumerate(annotations):

        file_save = file.split('.')[0]+'.txt'
        file_txt=os.path.join(outdir,file_save)
        f_w = open(file_txt,'w')

        in_file = open(file)
        tree=ET.parse(in_file)
        root = tree.getroot()
        size = root.find('size')
 
        w = int(size.find('width').text)
        h = int(size.find('height').text)
 
       # 在一个XML中每个Object的迭代
        for obj in root.iter('object'):
            # iter()方法可以递归遍历元素/树的所有子元素
            difficult = obj.find('difficult').text
            cls = obj.find('name').text
            # 如果训练标签中的品种不在程序预定品种，或者difficult = 1，跳过此object
            #if cls not in classes or int(difficult) == 1:
            #    continue
            #cls_id = classes.index(cls)#这里取索引，避免类别名是中文，之后运行yolo时要在cfg将索引与具体类别配对
            cls_id=0
            xmlbox = obj.find('bndbox')
     
            b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(
                xmlbox.find('ymin').text), float(xmlbox.find('ymax').text))
            bb = convert((w, h), b)
            f_w.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')

indir='./'
outdir='./'
xml_to_txt(indir,outdir)