YOLOv9图像标注和格式转换

一、软件安装

labelimg安装（anaconda）

方法一、

pip install labelImg

方法二、

pip install PyQt5 -i https://pypi.tuna.tsinghua.edu.cn/simple/
pip install pyqt5-tools -i https://pypi.tuna.tsinghua.edu.cn/simple/
pip install lxml -i https://pypi.tuna.tsinghua.edu.cn/simple/
pip install labelImg -i https://pypi.tuna.tsinghua.edu.cn/simple/

二、软件标注

打开labelimg环境,运行如下命令即可：

conda activate labelimg
labelimg

三、格式转换

标签文件.xml转换为.txt格式

import xml.etree.ElementTree as ET
import os
from os import getcwd

classes = ["pikaqiu",'apple']   # 改成自己的类别
abs_path = os.getcwd()
print(abs_path)

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(xml_id_path,txtpath):
    in_file = open(xml_id_path, encoding='UTF-8')     
    txtpath1=os.path.join(txtpath,xml_id_path.split('\\')[-1].split('.')[0]+'.txt')
    out_file = open(txtpath1, '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
        #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))
        b1, b2, b3, b4 = b

        # 标注越界修正
        if b2 > w:
            b2 = w
        if b4 > h:
            b4 = h
        b = (b1, b2, b3, b4)
        bb = convert((w, h), b)
        out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')


#xml标签文档地址          
xmlpath=r'C:\Users\admin\xxxxxxxx\labels_xml'
listxmls=os.listdir(xmlpath)
#txt文档保存地址
txtpath=r'C:\Users\admin\xxxxxxxx\labels_txt'
for xml_id in listxmls:
    xml_id_path=os.path.join(xmlpath,xml_id)
    convert_annotation(xml_id_path,txtpath)

四、训练数据参数设置

训练代码链接：项目目录预览 - yolov9 - GitCode

到yolov9主目录下的./data路径下重新编写一个.yaml文件

pikaqiu.yaml

train: /media/hadoop/xxxxxxx/VOC/images/train 
val: /media/hadoop/xxxxxxxxx/VOC/images/val
test: /media/hadoop/xxxxxxxx/VOC/images/test
# number of classes
nc: 2
# class names
names: [ 'pikaqiu','apple']

将该文件填入到训练文件train.py中的数据参数设置--data中: