(1)源代码地址:https://github.com/qqwweee/keras-yolo3
论文详解:https://blog.csdn.net/u014380165/article/details/80202337
直接安装相应的库即可实现,没有什么问题,只是看一下效果,后面没有用这个代码。
(2)训练自己的数据集:https://pjreddie.com/darknet/yolo/
1、下载darknet源代码安装,使用
下载源代码
git clone https://github.com/pjreddie/darknet
cd darknet
修改makeflie文件使用GPU,OpenCV
GPU=1
CUDNN=1
OpenCV=1
...
NVCC=/usr/local/cuda-8.0/bin/nvcc
编译
make -j7
清除
make clean
使用
wget https://pjreddie.com/media/files/yolov3.weights
./darknet detect cfg/yolov3.cfg yolov3.weights data/dog.jpg
如果出现GPU内存溢出,可以考虑将cfg/yolov3.cfg中的random改为0,batch和subdivisions改小
还不行可以考虑使用极小模型:
wget https://pjreddie.com/media/files/yolov3-tiny.weights
2、使用labelimg进行数据的标注:https://github.com/tzutalin/labelImg,生成.xml文件,还有一种标注工具是labelme
3、创建VOCdevkit 文件夹,具体格式参考;https://blog.csdn.net/john_bh/article/details/80625220
VOCdevkit
—VOC2007
——Annotations
——ImageSets
———Layout
———Main
———Segmentation
——JPEGImages
——labels
——my.py
Annotations中是所有的xml文件
JPEGImages中是所有的训练图片
Main中是4个txt文件,其中test.txt是测试集,train.txt是训练集,val.txt是验证集,trainval.txt是训练和验证集。
其他文件夹暂时可以不管
4、将“所有”图片放入JPEGImages中,将标注文件放入Annotations中,下面是其中my.py文件内容:
import os
import random
trainval_percent = 0.5
train_percent = 0.5
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(txtsavepath+'/trainval.txt', 'w')
ftest = open(txtsavepath+'/test.txt', 'w')
ftrain = open(txtsavepath+'/train.txt', 'w')
fval = open(txtsavepath+'/val.txt', 'w')
for i in list:
name=total_xml[i][:-4]+'\n'
if i in trainval:
ftrainval.write(name)
if i in train:
ftrain.write(name)
else:
fval.write(name)
else:
ftest.write(name)
ftrainval.close()
ftrain.close()
fval.close()
ftest .close()
运行python my.py即可生成ImageSets内的内容
5、在根目录下创建voc_label.py
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')]
classes = ["leaf","bottle","paomo"] #类别自己修改
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()
修改类别,执行
python voc_label.py
cat 2007_train.txt 2007_val.txt > train.txt
6、修改类别class,filters从下到上3处需要修改
[convolutional]
batch_normalize=1 ### BN
filters=32 ### 卷积核数目
size=3 ### 卷积核尺寸
stride=1 ### 卷积核步长
pad=1 ### pad
activation=leaky ### 激活函数
......
[convolutional]
size=1
stride=1
pad=1
filters=24 #9/3*(3+4+1)
activation=linear
[yolo]
mask = 6,7,8
anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326
classes=3 #类别
num=9
jitter=.3
ignore_thresh = .5
truth_thresh = 1
random=0 #1,如果显存很小,将random设置为0,关闭多尺度训练;
......
[convolutional]
size=1
stride=1
pad=1
filters=24 #9/3*(3+4+1)
activation=linear
[yolo]
mask = 3,4,5
anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326
classes=3 #类别
num=9
jitter=.3
ignore_thresh = .5
truth_thresh = 1
random=0 #1,如果显存很小,将random设置为0,关闭多尺度训练;
......
[convolutional]
size=1
stride=1
pad=1
filters=24 #9/3*(3+4+1)
activation=linear
[yolo]
mask = 0,1,2
anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326
classes=3
num=9
jitter=.3
ignore_thresh = .5
truth_thresh = 1
random=0 #1,如果显存很小,将random设置为0,关闭多尺度训练;
如果训练显存不足,可以考虑使用yolov3-tiny.cfg,做同样的修改,类别class,filters从下到上2处需要修改
7、修改data/voc.names,设置成自己的类别,例如:
leaf
bottle
paomo
8、修改cfg/voc.data,如下:
classes= 3
train = ~/study/depthlearning/darknet/train.txt
valid = ~/study/depthlearning/darknet/2007_test.txt
names = data/voc.names
backup = backup
9、训练
wget https://pjreddie.com/media/files/darknet53.conv.74
普通模型:
如果只有一个显卡:-gpus 0,两个则-gpus 0,1
./darknet detector train cfg/voc.data cfg/yolov3-voc.cfg darknet53.conv.74 -gpus 0
极小模型:
./darknet detector train cfg/voc.data cfg/yolov3-tiny.cfg -gpus 0
10:测试
图片
普通:
./darknet detector test cfg/voc.data cfg/yolov3-voc.cfg backup/yolov3-voc_900.weights 1.jpg
极小:
./darknet detector test cfg/voc.data cfg/yolov3-tiny.cfg backup/yolov3-tiny_20000.weights 1.jpg
视频
普通:
./darknet detector demo cfg/voc.data cfg/yolov3-voc.cfg backup/yolov3-voc_900.weights test.avi
极小:
./darknet detector demo cfg/voc.data cfg/yolov3-tiny.cfg backup/yolov3-tiny_20000.weights test.avi
参考博客:https://blog.csdn.net/john_bh/article/details/80625220
https://blog.csdn.net/just_sort/article/details/81389571
https://blog.csdn.net/helloworld1213800/article/details/79749359