如何使用YOLOv8训练自己的实例分割模型

1. 准备数据集

1.1 将coco json格式的标签文件转换为.txt格式

准备文件夹如下：

而VOCdevkit文件夹中如下

json2txt.py文件中放入以下代码并运行。

import json
import os
import argparse
from tqdm import tqdm

def convert_label_json(json_dir, save_dir, class_id_mapping):
    json_paths = os.listdir(json_dir)

    for json_path in tqdm(json_paths):
        path = os.path.join(json_dir, json_path)
        with open(path, 'r') as load_f:
            json_dict = json.load(load_f)

        image_id_to_annotations = {}  # Collect annotations per image
        for annotation in json_dict['annotations']:
            image_id = annotation['image_id']
            if image_id not in image_id_to_annotations:
                image_id_to_annotations[image_id] = []
            image_id_to_annotations[image_id].append(annotation)

        for image_info in json_dict['images']:
            image_id = image_info['id']
            h, w = image_info['height'], image_info['width']

            annotations = image_id_to_annotations.get(image_id, [])
            if len(annotations) == 0:
                continue

            txt_filename = image_info['file_name'].replace('.jpg', '.txt')
            txt_path = os.path.join(save_dir, txt_filename)
            txt_file = open(txt_path, 'w')

            for annotation in annotations:
                category_id = annotation['category_id']
                label_index = class_id_mapping.get(category_id, -1)
                if label_index == -1:
                    continue

                segmentation = annotation['segmentation'][0]  # Get the first segmentation polygon
                normalized_segmentation = [coord / w if idx % 2 == 0 else coord / h for idx, coord in enumerate(segmentation)]
                normalized_segmentation_str = ' '.join(map(str, normalized_segmentation))
                label_str = f"{label_index} {normalized_segmentation_str}\n"
                txt_file.write(label_str)
            txt_file.close()

if __name__ == "__main__":
    parser = argparse.ArgumentParser(description='json convert to txt params')
    parser.add_argument('--json-dir', type=str, default=r"VOCdevkit\json", help='json path dir')
    parser.add_argument('--save-dir', type=str, default=r"VOCdevkit\txt", help='txt save dir')
    args = parser.parse_args()

    class_id_mapping = {1: 0}  # Mapping category_id to label index
    json_dir = args.json_dir
    save_dir = args.save_dir
    convert_label_json(json_dir, save_dir, class_id_mapping)

1.2 划分训练集、验证集、测试集

split_data.py文件中放入以下代码并运行。

# 将图片和标注数据按比例切分为 训练集和测试集
import shutil
import random
import os
import argparse


# 检查文件夹是否存在
def mkdir(path):
    if not os.path.exists(path):
        os.makedirs(path)


def main(image_dir, txt_dir, save_dir):
    # 创建文件夹
    mkdir(save_dir)
    images_dir = os.path.join(save_dir, 'images')
    labels_dir = os.path.join(save_dir, 'labels')

    img_train_path = os.path.join(images_dir, 'train')
    img_test_path = os.path.join(images_dir, 'test')
    img_val_path = os.path.join(images_dir, 'val')

    label_train_path = os.path.join(labels_dir, 'train')
    label_test_path = os.path.join(labels_dir, 'test')
    label_val_path = os.path.join(labels_dir, 'val')

    mkdir(images_dir);
    mkdir(labels_dir);
    mkdir(img_train_path);
    mkdir(img_test_path);
    mkdir(img_val_path);
    mkdir(label_train_path);
    mkdir(label_test_path);
    mkdir(label_val_path);

    # 数据集划分比例，训练集80%，验证集10%，测试集10%，按需修改
    train_percent = 0.8
    val_percent = 0.1
    test_percent = 0.1

    total_txt = os.listdir(txt_dir)
    num_txt = len(total_txt)
    list_all_txt = range(num_txt)  # 范围 range(0, num)

    num_train = int(num_txt * train_percent)
    num_val = int(num_txt * val_percent)
    num_test = num_txt - num_train - num_val

    train = random.sample(list_all_txt, num_train)
    # 在全部数据集中取出train
    val_test = [i for i in list_all_txt if not i in train]
    # 再从val_test取出num_val个元素，val_test剩下的元素就是test
    val = random.sample(val_test, num_val)

    print("训练集数目：{}, 验证集数目：{},测试集数目：{}".format(len(train), len(val), len(val_test) - len(val)))
    for i in list_all_txt:
        name = total_txt[i][:-4]

        srcImage = os.path.join(image_dir, name + '.jpg')
        srcLabel = os.path.join(txt_dir, name + '.txt')

        if i in train:
            dst_train_Image = os.path.join(img_train_path, name + '.jpg')
            dst_train_Label = os.path.join(label_train_path, name + '.txt')
            shutil.copyfile(srcImage, dst_train_Image)
            shutil.copyfile(srcLabel, dst_train_Label)
        elif i in val:
            dst_val_Image = os.path.join(img_val_path, name + '.jpg')
            dst_val_Label = os.path.join(label_val_path, name + '.txt')
            shutil.copyfile(srcImage, dst_val_Image)
            shutil.copyfile(srcLabel, dst_val_Label)
        else:
            dst_test_Image = os.path.join(img_test_path, name + '.jpg')
            dst_test_Label = os.path.join(label_test_path, name + '.txt')
            shutil.copyfile(srcImage, dst_test_Image)
            shutil.copyfile(srcLabel, dst_test_Label)


if __name__ == '__main__':
    """
    python split_datasets.py --image-dir my_datasets/color_rings/imgs --txt-dir my_datasets/color_rings/txts --save-dir my_datasets/color_rings/train_data
    """
    parser = argparse.ArgumentParser(description='split datasets to train,val,test params')
    parser.add_argument('--image-dir', type=str, default=r"VOCdevkit\images", help='image path dir')
    parser.add_argument('--txt-dir', type=str, default=r"VOCdevkit\txt", help='txt path dir')
    parser.add_argument('--save-dir', default=r"VOCdevkit\Dataset", type=str, help='save dir')
    args = parser.parse_args()
    image_dir = args.image_dir
    txt_dir = args.txt_dir
    save_dir = args.save_dir

    main(image_dir, txt_dir, save_dir)

运行过后会在Dataset文件夹中产生如下子文件夹，文件夹中存放着对应内容。至此数据集准备完毕。

2. 新建data.yaml文件

放入以下内容。并修改文件路径和class name

# Ultralytics YOLO 🚀, AGPL-3.0 license
# COCO 2017 dataset http://cocodataset.org by Microsoft
# Example usage: yolo train data=coco.yaml
# parent
# ├── ultralytics
# └── datasets
#     └── coco  ← downloads here (20.1 GB)


# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
path: \split  # dataset root dir
train: \split\images\train  # train images (relative to 'path') 118287 images
val: \split\images\val  # val images (relative to 'path') 5000 images
test: \split\images\test  # 20288 of 40670 images, submit to https://competitions.codalab.org/competitions/20794

# Classes
names:
  0: pothole

3. 修改default.yaml文件

找到YOLO v8\ultralytics-main\ultralytics/yolo/cfg文件路径下的default.yaml文件并打开

task设置为segment；

mode设置为train；

model后设置为预训练模型yolov8n-seg.pt的文件路径；

预训练模型需到官网下载 https://docs.ultralytics.com/tasks/segment/

data后设置为《2.新建data.yaml文件》中所述的data.yaml文件的绝对路径；

epoch设置为自己所需的训练轮数；

batch根据自己电脑配置进行设置