1 首先文件夹目录如下排列,
Annotations里面装的是数据的标签文件(xml格式)
JEPGImages装的是对应的图片
ImageSets文件装的是Main,Main下面就是用来装train.txt, test.txt, trainval.txt , val.txt.
train.txt里面的内容是要训练的图片名字,所以test.txt也就是用来测试的图片名字
2 但是这还不是yolo3输入的数据格式,我们要在生成一个 图片与标签在一起的txt文件,这个txt文件像下图这种的
由于截屏宽度有限,故第一行完整的是这样的
E:\Jupyter Notebook\demo\defect_detection\yolo3-keras\yolo3-keras-master\yolo3-keras-master/VOCdevkit/VOCPCB/JPEGImages/l_light_11_short_06_1_600.jpg 396,520,422,553,3 516,516,543,550,3
如何生成,使用以下代码
import xml.etree.ElementTree as ET
from os import getcwd
sets=[('PCB', 'train'), ('PCB', 'val'), ('PCB', 'test')]
classes = ["missing_hole", "mouse_bite", "open_circuit", "short", "spur", "spurious_copper"]
def convert_annotation(year, image_id, list_file):
in_file = open('VOCdevkit/VOC%s/Annotations/%s.xml'%(year, image_id))
tree=ET.parse(in_file)
root = tree.getroot()
for obj in root.iter('object'):
difficult = obj.find('difficult').text
cls1 = obj.find('name').text
if cls1 not in classes or int(difficult)==1:
continue
cls_id = classes.index(cls1)
xmlbox = obj.find('bndbox')
b = (int(xmlbox.find('xmin').text), int(xmlbox.find('ymin').text), int(xmlbox.find('xmax').text), int(xmlbox.find('ymax').text))
list_file.write(" " + ",".join([str(a) for a in b]) + ',' + str(cls_id))
wd = getcwd()
for year, image_set in sets:
image_ids = open('VOCdevkit/VOC%s/ImageSets/Main/%s.txt'%(year, image_set)).read().strip().split()
list_file = open('%s_%s.txt'%(year, image_set), 'w')
for image_id in image_ids:
list_file.write('%s/VOCdevkit/VOC%s/JPEGImages/%s.jpg'%(wd, year, image_id))
convert_annotation(year, image_id, list_file)
list_file.write('\n')
list_file.close()