1、下载VOC2007数据集,把所有文件夹里面的东西删除,保留所有文件夹的名字(或者参照VOC2007新建文件夹)。如下图所示:
2、把所有的图片都复制到JPEGImages文件夹中。
3、生成Annotations下的文件
3.1、用labelimg标注图片生成xml数据
使用标注工具labelimg手动给每一张图片加标签生成xml格式文件并放入Annotations文件夹中。
本人使用的是ImageNet数据集中的部分图片并自己填充一些图片。
3.2、下载Iabelimg(https://github.com/tzutalin/labelImg)
sudo apt-get install pyqt4-dev-tools # 安装PyQt4
sudo pip install lxml (# 安装lxml,如果报错,可以试试下面语句sudo apt-get install python-lxml)
git clone https://github.com/tzutalin/labelImg.git
cd labelImg
make all
./labelImg.py
使用labelimg给图片标注,标注结果(xml文件)保存在Annotations文件夹中。
3.3、将xml转换成为darknet能读入的txt格式文件
3.31、生成ImageSets/Main/4个文件
在VOC2007下建个文件test.py,并运行,结果如下:
test.py代码如下:
import os
import random
trainval_percent = 0.1
train_percent = 0.9
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('ImageSets/Main/trainval.txt', 'w')
ftest = open('ImageSets/Main/test.txt', 'w')
ftrain = open('ImageSets/Main/train.txt', 'w')
fval = open('ImageSets/Main/val.txt', 'w')
for i in list:
name = total_xml[i][:-4] + '\n'
if i in trainval:
ftrainval.write(name)
if i in train:
ftest.write(name)
else:
fval.write(name)
else:
ftrain.write(name)
ftrainval.close()
ftrain.close()
fval.close()
ftest.close()
3.32创建文件名为voc_label.py的文件
复制以下代码(记得将路径改成自己的,把所有#后面的注释删掉,否则运行不了)
import xml.etree.ElementTree as ET
import pickle
import os
from os import listdir, getcwd
from os.path import join
sets=[('2007', 'train'), ('2007', 'val'), ('2007', 'test')]
#按需修改
classes = ["garbabe can","clothes","carton","cell phone","bottle","cup","file","towel","spoon","roll paper","bed quilt","bowl","plate","window","book","fruit"]
#改为自己的类别
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 convert_annotation(year, image_id):
in_file = open('VOCdevkit/VOC%s/Annotations/%s.xml'%(year, image_id))
out_file = open('VOCdevkit/VOC%s/labels/%s.txt'%(year, image_id), 'w')
tree=ET.parse(in_file)
root = tree.getroot()
size = root.find('size')
w = int(size.find('width').text)
h = int(size.find('height').text)
for obj in root.iter('object'):
difficult = obj.find('difficult').text
cls = obj.find('name').text
if cls not in classes or int(difficult) == 1:
continue
cls_id = classes.index(cls)
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)
out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')
wd = getcwd()
for year, image_set in sets:
if not os.path.exists('VOCdevkit/VOC%s/labels/'%(year)):
os.makedirs('VOCdevkit/VOC%s/labels/'%(year))
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\n'%(wd, year, image_id))
convert_annotation(year, image_id)
list_file.close()
运行上述文件:
python voc_lable.py
在VOC2007文件夹下生成labels文件夹,labels文件夹内存储由xml格式标注文件转化生成的TXT格式文件。同时在darknet/VOCdevkit下生成2007_test.txt、2007_train.txt、2007_val.txt文件。2007_train.txt文件中包含要训练的图片以及图片的路径。
3.4修改配置文件
1、创建cfg/lsq.data,写入如下代码,并根据自己的类别数量和路径修改文件。
classes=16
train=/home/lsq/darknet/VOCdevkit/2007_train.txt
valid=/home/lsq/darknet/VOCdevkit/2007_test.txt
names=data/lsq.name
backup=backup
2、复制cfg/yolov3-voc.cfg并重命名yolov3-lsq.cfg,然后修改如下内容:[yolo]下面的classes=16为自己的类别数,[yolo]上面的[convolutional]中的filters=63为3*(5+classes),共需要修改3处。
3、创建data/lsq.name,将自己训练的类别写入。
4、本人使用GPU、CUDNN、OPENCV,因此修改darknet文件下的Makefile如下:
5、参考本人另一篇博客:https://editor.csdn.net/md/?articleId=108362959
4 训练&测试
1、训练
./darknet detector train cfg/lsq.data cfg/yolov3-lsq.cfg
2、假如训练900次时想要测试以下训练效果
./darknet detect cfg/yolov3-lsq.cfg backup/yolov3-lsq_900.weights prediction.jpg -thresh 0.4
3、从中止处继续训练
./darknet detector train cfg/lsq.data cfg/yolov3-lsq.cfg backup/yolov3-lsq_900.weights
训练的权重文件保存在darknet/backup文件夹中。