pennFudanPed数据集只有不到200张图片,标注的行人为非遮挡行人,遮挡和骑车的行人都没有标注
标注框做的比较标注和准确,下面是该数据集直接转成yolo格式的代码,新建train文件夹,然后使用下面代码存为py,然后运行
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import os
import re
from pathlib import Path
from PIL import Image
import csv
import shutil
import numpy as np
pngpath='./PNGImages/'
oritxt='./Annotation/'
newtxt='./train/'
savepath='./train/'
def convert(size, box0,box1,box2,box3):
dw = 1. / size[0]
dh = 1. / size[1]
x = (box0 + box2) / 2 * dw
y = (box1 + box3)/ 2 * dh
w = (box2-box0) * dw
h = (box3-box1) * dh
return (x, y, w, h)
txtfiles = os.listdir(oritxt)
matrixs = []
for txtpath in txtfiles:
img_path = pngpath + txtpath[:-4] + ".png";
outpath = newtxt + txtpath[:-4] + '.txt'
with Image.open(img_path) as Img:
img_size = Img.size
ans = ''
f1 = open(oritxt + txtpath, 'r')
for line in f1.readlines():
if re.findall('Xmin',line):
pt = [int(x) for x in re.findall(r"\d+",line)]
matrixs.append(pt)
bb = convert(img_size, pt[1], pt[2], pt[3], pt[4])
ans = ans + '1' + ' ' + ' '.join(str(a) for a in bb) + '\n'
f1.close()
with open(outpath, 'w') as outfile:
outfile.write(ans)
shutil.copy(img_path, savepath + txtpath[:-4] + '.png')