visdrone+yolov5

1.数据集转换
VisDrone2019-DET-train
下载地址

http://aiskyeye.com/download/object-detection/

解压后只有images、annotations两个文件夹
使用数据集转换工具visdrone2yolo.py

只要修改root_dir就可以直接运行。运行前先装一下pip包。如果没有labels文件夹生成就手动新建个labels文件夹，在root_dir下。

import os
from os import getcwd
from PIL import Image
import xml.etree.ElementTree as ET
import random

#root_dir = "train/"
root_dir = "/home/lhw/Gradute/VirDrone/yolov5/VisDrone2019-DET-train/"
annotations_dir = root_dir+"annotations/"
image_dir = root_dir + "images/"
label_dir = root_dir + "labels/"
# label_dir = root_dir + "images/"    # yolo里面要和图片放到一起
xml_dir = root_dir+"annotations_voc/"  #注意新建文件夹。后续改一下名字，运行完成之后annotations这个文件夹就不需要了。把annotations_命名为annotations
data_split_dir = root_dir + "train_namelist/"

sets = ['train', 'test','val']
class_name = ['ignored regions', 'pedestrian','people','bicycle','car', 'van', 'truck', 'tricycle','awning-tricycle', 'bus','motor','others']


def visdrone2voc(annotations_dir, image_dir, xml_dir):
    for filename in os.listdir(annotations_dir):
        fin = open(annotations_dir + filename, 'r')
        image_name = filename.split('.')[0]
        img = Image.open(image_dir + image_name + ".jpg")
        xml_name = xml_dir + image_name + '.xml'
        with open(xml_name, 'w') as fout:
            fout.write('' + '\n')

            fout.write('\t' + 'VOC2007' + '\n')
            fout.write('\t' + '' + image_name + '.jpg' + '' + '\n')

            fout.write('\t' + '' + '\n')
            fout.write('\t\t' + '' + 'VisDrone2018 Database' + '' + '\n')
            fout.write('\t\t' + '' + 'VisDrone2018' + '' + '\n')
            fout.write('\t\t' + '' + 'flickr' + '' + '\n')
            fout.write('\t\t' + '' + 'Unspecified' + '' + '\n')
            fout.write('\t' + '' + '\n')

            fout.write('\t' + '' + '\n')
            fout.write('\t\t' + '' + 'Haipeng Zhang' + '' + '\n')
            fout.write('\t\t' + '' + 'Haipeng Zhang' + '' + '\n')
            fout.write('\t' + '' + '\n')

            fout.write('\t' + '' + '\n')
            fout.write('\t\t' + '' + str(img.size[0]) + '' + '\n')
            fout.write('\t\t' + '' + str(img.size[1]) + '' + '\n')
            fout.write('\t\t' + '' + '3' + '' + '\n')
            fout.write('\t' + '' + '\n')

            fout.write('\t' + '' + '0' + '' + '\n')

            for line in fin.readlines():
                line = line.split(',')
                fout.write('\t' + '' + '\n')

            fin.close()
            fout.write('')

def data_split(xml_dir, data_split_dir):
    trainval_percent = 0.2
    train_percent = 0.9
    total_xml = os.listdir(xml_dir)
    if not os.path.exists(data_split_dir):
        os.makedirs(data_split_dir)
    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(data_split_dir+'/trainval.txt', 'w')
    ftest = open(data_split_dir+'/test.txt', 'w')
    ftrain = open(data_split_dir+'/train.txt', 'w')
    fval = open(data_split_dir+'/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()


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_voc(xml_dir, label_dir, image_name):
    in_file = open(xml_dir + '%s.xml' % (image_name))
    out_file = open(label_dir + '%s.txt' % (image_name), '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 class_name or int(difficult) == 1:
            continue
        cls_id = class_name.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)
        if cls_id != 0:  # 忽略掉0类
            if cls_id != 11:  # 忽略掉11类
                out_file.write(str(cls_id - 1) + " " + " ".join([str(a) for a in bb]) + '\n')  # 其他类id-1。可以根据自己需要修改代码

def voc2yolo(xml_dir, image_dir, label_dir):
    wd = getcwd()
    print(wd)
    for image_set in sets:
        if not os.path.exists(label_dir):
            os.makedirs(label_dir)
        image_names = open(data_split_dir+'%s.txt' % (image_set)).read().strip().split()
        list_file = open(root_dir + '%s.txt' % (image_set), 'w')
        for image_name in image_names:
            list_file.write(image_dir+'%s.jpg\n' % (image_name))
            convert_annotation_voc(xml_dir, label_dir, image_name)
        list_file.close()




if __name__ == '__main__':
    visdrone2voc(annotations_dir, image_dir, xml_dir) #将visdrone转化为voc的xml格式
    data_split(xml_dir, data_split_dir)		# 将数据集分开成train、val、test
    voc2yolo(xml_dir, image_dir, label_dir)	# 将voc转化为yolo格式的txt

2. 下载yolov5源码

git clone https://github.com/ultralytics/yolov5.git

然后在git下的yolov5根目录下创建文件夹visdronedata及其附属目录

将visdrone的images文件夹里面的图片全部复制到images/train和iamges/val里面，上面程序生成的labels文件夹，将里面的所有txt复制到labels/train和labels/val里面

然后安装yolov5环境，要在Python3.8的环境下安装，如果使用conda可以使用

conda create -n py38 python=3.8   # 创建虚拟环境
conda activate py38 		# 激活虚拟环境，如果成功激活环境则在命令行用户名前面有虚拟环境名称的括号
pip install -r requirements.txt # 在yolov5根目录下安装需要的包

安装完环境之后，修改data/voc.yaml

# download command/URL (optional)
# download: bash data/scripts/get_voc.sh

# train and val data as 1) directory: path/images/, 2) file: path/images.txt, or 3) list: [path1/images/, path2/images/]
train: visdronedata/images/train/  # 16551 images
val: visdronedata/images/train/  # 4952 images

# number of classes
nc: 10

# class names
names: ['pedestrian','people','bicycle','car','van','truck','tricycle','awning-tricycle','bus','motor']

修改models/yolov5l.yaml

# parameters
nc: 10  # number of classes   #只修改这个类别数
depth_multiple: 1.0  # model depth multiple
width_multiple: 1.0  # layer channel multiple

在这里的v1部分下载预训练模型yolov5l.pt，由于使用的yolov5l.yaml，所以使用yolov5l.pt。使用其他的训练配置文件可以下载相应的预训练模型，在github有对应关系。
将下载好的预训练模型放到weights文件夹下

然后在命令段运行

内存足够时可以增大batch-size：1、8、16、32、64

python train.py --data data/voc.yaml --cfg models/yolov5l.yaml --weights weights/yolov5l.pt --batch-size 1

也可直接在train.py直接修改默认参数，然后直接运行

python train.py

然后进行训练，在训练的过程中会产生过程文件及训练模型，会保存在runs/文件夹中

在exp里面会有保存的中间临时权重，可以拿出来放到yolov5/根目录下进行预测测试，预测的结果也会放到runs/detect文件夹下

python decect.py --source file.jpg --weight best.pt --conf 0.25
python decect.py --source file.mp4 --weight best.pt --conf 0.25

运行十来分钟就拿来测试了。。。