程序员

YOLOv3训练并检测自己的数据 2020

2020-03-08  本文已影响0人  绍重先

1 配置环境

本机环境

系统:Windows10

Python:3
IDE:VS2017

nvcc -V

会出现版本信息

编译需添加OpenCV环境依赖

VC++ 目录—>包含目录—>编辑,添加以下三项[选择自己安装位置的绝对路径]

C:\opencv\build\include
C:\opencv\build\include\opencv
C:\opencv\build\include\opencv2

VC++ 目录—>库目录中添加

C:\opencv\build\x64\vc15\lib

链接器->输入->附加依赖项添加[根据自己安装的版本]
opencv_world342d.lib
opencv_world342.lib

编译完成后,darknet.exe会在x64文件夹中

2 数据集准备

2.1以VOC格式准备自己的数据集文件夹

├─VOCdevkit2007
│  └─VOC2007
│      ├─Annotations
│      ├─ImageSets
│      │  └─Main
│      ├─JPEGImages
│      └─labels

2.2使用脚本批量更改图片名称

import os
path = r'C:\Users\Dexter0ion\Desktop\TrainData\VOCdevkit2007\VOC2007\JPEGImages'
filelist = os.listdir(path) # 该文件夹下所有的文件(包括文件夹)
count=0 # 编号从0开始

for file in filelist:
    print(file)

for file in filelist:  
# 遍历所有文件
    Olddir=os.path.join(path,file)   # 原来的文件路径
    if os.path.isdir(Olddir):   # 如果是文件夹则跳过
        continue
    filename=os.path.splitext(file)[0]   # 文件名
    filetype=os.path.splitext(file)[1]   # 文件扩展名
    Newdir=os.path.join(path,str(count).zfill(6)+filetype)  # 用字符串函数zfill 以0补全所需位数
    os.rename(Olddir,Newdir) # 重命名
    count+=1

运行命令

python ./rename.py

2.3使用labelImg软件对数据进行标注

labelImg下载地址:http://tzutalin.github.io/labelImg/

解压后在data/predefined_classes.txt中修改预设的class名字

即可开始标注,快捷键流程

[w]框选
[Ctrl+s]保存
[d]下一张

3.处理标注后的数据

3.1生成Main目录下的txt文件

import os
import random

trainval_percent = 0.7   # trainval占总数的比例
train_percent = 0.5   # train占trainval的比例
xmlfilepath = r'C:\Users\Dexter0ion\Desktop\TrainData\VOCdevkit2007\VOC2007\Annotations'
txtsavepath = r'C:\Users\Dexter0ion\Desktop\TrainData\VOCdevkit2007\VOC2007\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 + r'\trainval.txt', 'w')
ftest = open(txtsavepath + r'\test.txt', 'w')
ftrain = open(txtsavepath + r'\train.txt', 'w')
fval = open(txtsavepath + r'\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()
运行命令
python ./generatetxt.py

3.2生成darknet可用的yolo类型数据

将VOCdevkit2007文件夹整个复制到

darknet-master\darknet-master\build\darknet

文件夹下

进入
darknet-master\darknetmaster\build\darknet\VOCdevkit2007
文件夹
创建voc_label.py脚本

voc_label.py[此次只训练一个目标,在classes中改为你要训练的目标名字,多个则用逗号分隔]

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 = ["redbox"]


def convert(size, box):
    dw = 1./(size[0])
    dh = 1./(size[1])
    x = (box[0] + box[1])/2.0 - 1
    y = (box[2] + box[3])/2.0 - 1
    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('VOC%s/Annotations/%s.xml'%(year, image_id))
    out_file = open('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('VOC%s/labels/'%(year)):
        os.makedirs('VOC%s/labels/'%(year))
    image_ids = open('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/VOC%s/JPEGImages/%s.jpg\n'%(wd, year, image_id))
        convert_annotation(year, image_id)
    list_file.close()

os.system("cat 2007_train.txt 2007_val.txt 2012_train.txt 2012_val.txt > train.txt")
os.system("cat 2007_train.txt 2007_val.txt 2007_test.txt 2012_train.txt 2012_val.txt > train.all.txt")


之后会在VOC2007同目录下得到

│  2007_test.txt
│  2007_train.txt
│  2007_val.txt

4 训练准备

需要配置

下载地址:http://pjreddie.com/media/files/darknet53.conv.74
下载后移动到:darknet-master\build\darknet\x64
文件夹下
classes= 1
#自己先前生成文件的绝对路径
train  = C:\Users\Dexter0ion\Desktop\TrainData\darknet-master\darknet-master\build\darknet\VOCdevkit2007\2007_train.txt  
valid  = C:\Users\Dexter0ion\Desktop\TrainData\darknet-master\darknet-master\build\darknet\VOCdevkit2007\2007_test.txt  
names = data/obj.names
backup = backup/
redbox

修改batchsubdivisions

修改max_batches(作者声明最好是2000*训练目标个数,但不要小于4000)和steps(80%,90%,降低学习率阈值)

修改三处[convolutional] [yolo]

filters = 3*(5+classes数目) classes = 本次训练目标数,即1个



5 开始训练

darknet-master\build\darknet\x64目录下运行指令

./darknet.exe detector train data/obj.data yolo-obj.cfg darknet53.conv.74 mjpeg_port 8090 -ext_output  | Out-File ./alpha_train_log.txt

注意:前100次loss会很高,之后会逐步下降

训练完成的权重文件默认保存在backup文件夹中

6 测试训练结果

复制yolo-obj.cfg且重命名为yolo-obj-test.cfg

darknet-master\build\darknet\x64目录下运行指令

./darknet detector test data/obj.data yolo-obj-test.cfg backup/yolo-obj_last.weights -thresh 0.1
thresh 为0,过低的结果 正常的结果
上一篇 下一篇

猜你喜欢

热点阅读