MMdetection自定义数据集训练及相关配置

MMdetection自定义数据集训练及相关配置

一、安装mmdetection

pip install -U openmim
mim install mmengine
mim install "mmcv>=2.0.0"
git clone https://github.com/open-mmlab/mmdetection.git
cd mmdetection
pip install -v -e .

安装完以后，验证一下是否安装正确。

mim download mmdet --config rtmdet_tiny_8xb32-300e_coco --dest .
python demo/image_demo.py demo/demo.jpg rtmdet_tiny_8xb32-300e_coco.py --weights rtmdet_tiny_8xb32-300e_coco_20220902_112414-78e30dcc.pth --device cuda:0

如果不报错，有正常结果，代表安装成功。

其他的包可以通过

pip install -r requirements.txt

注意：中间遇到缺少的库自己安装，比如pytorch，根据自己的硬件环境安装对应的pytorch版本。

二、数据准备

MMdetection内置的模型大多为coco数据集格式，voc格式的模型较少。建议直接使用coco模式，voc模式有很多坑，

将原始数据集转换为coco格式。首先将图像和xml文件都拷贝到同一个文件夹当中，执行以下脚本得到coco数据集。

# coding:utf-8
# pip install lxml

import os
import glob
import json
import shutil
import numpy as np
import xml.etree.ElementTree as ET

path2 = "."

START_BOUNDING_BOX_ID = 1


def get(root, name):
    return root.findall(name)


def get_and_check(root, name, length):
    vars = root.findall(name)
    if len(vars) == 0:
        raise NotImplementedError('Can not find %s in %s.' % (name, root.tag))
    if length > 0 and len(vars) != length:
        raise NotImplementedError('The size of %s is supposed to be %d, but is %d.' % (name, length, len(vars)))
    if length == 1:
        vars = vars[0]
    return vars


def convert(xml_list, json_file):
    json_dict = {"images": [], "type": "instances", "annotations": [], "categories": []}
    categories = pre_define_categories.copy()
    bnd_id = START_BOUNDING_BOX_ID
    all_categories = {}
    for index, line in enumerate(xml_list):
        # print("Processing %s"%(line))
        xml_f = line
        tree = ET.parse(xml_f)
        root = tree.getroot()

        filename = os.path.basename(xml_f)[:-4] + suffix
        image_id = index +1
        size = get_and_check(root, 'size', 1)
        width = int(get_and_check(size, 'width', 1).text)
        height = int(get_and_check(size, 'height', 1).text)
        image = {'file_name': filename, 'height': height, 'width': width, 'id': image_id}
        json_dict['images'].append(image)
        ## Cruuently we do not support segmentation
        #  segmented = get_and_check(root, 'segmented', 1).text
        #  assert segmented == '0'
        for obj in get(root, 'object'):
            category = get_and_check(obj, 'name', 1).text
            if category in all_categories:
                all_categories[category] += 1
            else:
                all_categories[category] = 1
            if category not in categories:
                if only_care_pre_define_categories:
                    continue
                new_id = len(categories) + 1
                print(
                    "[warning] category '{}' not in 'pre_define_categories'({}), create new id: {} automatically".format(
                        category, pre_define_categories, new_id))
                categories[category] = new_id
            category_id = categories[category]
            bndbox = get_and_check(obj, 'bndbox', 1)
            xmin = int(float(get_and_check(bndbox, 'xmin', 1).text))
            ymin = int(float(get_and_check(bndbox, 'ymin', 1).text))
            xmax = int(float(get_and_check(bndbox, 'xmax', 1).text))
            ymax = int(float(get_and_check(bndbox, 'ymax', 1).text))
            assert (xmax > xmin), "xmax <= xmin, {}".format(line)
            assert (ymax > ymin), "ymax <= ymin, {}".format(line)
            o_width = abs(xmax - xmin)
            o_height = abs(ymax - ymin)
            ann = {'area': o_width * o_height, 'iscrowd': 0, 'image_id':
                image_id, 'bbox': [xmin, ymin, o_width, o_height],
                   'category_id': category_id, 'id': bnd_id, 'ignore': 0,
                   'segmentation': []}
            json_dict['annotations'].append(ann)
            bnd_id = bnd_id + 1

    for cate, cid in categories.items():
        cat = {'supercategory': 'none', 'id': cid, 'name': cate}
        json_dict['categories'].append(cat)
    json_fp = open(os.path.join("annotations", json_file), 'w')
    json_str = json.dumps(json_dict)
    json_fp.write(json_str)
    json_fp.close()
    print("------------create {} done--------------".format(json_file))
    print("find {} categories: {} -->>> your pre_define_categories {}: {}".format(len(all_categories),
                                                                                  all_categories.keys(),
                                                                                  len(pre_define_categories),
                                                                                  pre_define_categories.keys()))
    print("category: id --> {}".format(categories))
    print(categories.keys())
    print(categories.values())


if __name__ == '__main__':
    # 类别
    classes = ['water']
    # 后缀
    suffix = '.jpg'
    pre_define_categories = {}
    for i, cls in enumerate(classes):
        pre_define_categories[cls] = i + 1
    # pre_define_categories = {'a1': 1, 'a3': 2, 'a6': 3, 'a9': 4, "a10": 5}
    only_care_pre_define_categories = True
    # only_care_pre_define_categories = False

    train_ratio = 0.9
    save_json_train = 'instances_train2017.json'
    save_json_val = 'instances_val2017.json'
    xml_dir = "./imgAndXml"

    xml_list = glob.glob(xml_dir + "/*.xml")
    xml_list = np.sort(xml_list)
    np.random.seed(100)
    np.random.shuffle(xml_list)

    train_num = int(len(xml_list) * train_ratio)
    xml_list_train = xml_list[:train_num]
    xml_list_val = xml_list[train_num:]

    if os.path.exists(path2 + "/annotations"):
        shutil.rmtree(path2 + "/annotations")
    os.makedirs(path2 + "/annotations")

    convert(xml_list_train, save_json_train)
    convert(xml_list_val, save_json_val)


    if os.path.exists(path2 + "/train2017"):
        shutil.rmtree(path2 + "/train2017")
    os.makedirs(path2 + "/train2017")
    if os.path.exists(path2 + "/val2017"):
        shutil.rmtree(path2 + "/val2017")
    os.makedirs(path2 + "/val2017")


    f1 = open("train.txt", "w")
    for xml in xml_list_train:
        img = xml[:-4] + suffix
        f1.write(os.path.basename(xml)[:-4] + "\n")
        shutil.copyfile(img, path2 + "/train2017/" + os.path.basename(img))

    f2 = open("test.txt", "w")
    for xml in xml_list_val:
        img = xml[:-4] + suffix
        f2.write(os.path.basename(xml)[:-4] + "\n")
        shutil.copyfile(img, path2 + "/val2017/" + os.path.basename(img))
    f1.close()
    f2.close()

    os.remove("train.txt")
    os.remove("test.txt")

    print("-------------------------------")
    print("train number:", len(xml_list_train))
    print("val number:", len(xml_list_val))

将所有的图像和xml文件放在了imgAndXml文件夹当中。其中红色为事先准备的文件信息，绿色框为后面生成的，annotations、train2017、val2017为生成的coco数据集，也就是转换之后的，目录如下图所示。

三、修改配置文件

基础修改

mmdetection/mmdet/datasets/coco.py 修改，修改classes（只有一个时需要加个逗号）和palette（标签颜色）

mmdetection/mmdet/evaluation/functional/class_names.py

注：上面的修改完成之后需要编译安装，才生效

python setup.py install

下面的网络结构文件num_classes同样需要修改，但不需要编译，对应训练的类别，和上面保持一致即可,如下图：
mmdetection/configs/_base_/datasets/faster-rcnn_r50_fpn.py

进阶修改

修改epoch、学习率

mmdetection/configs/_base_/schedules//schedule_1x.py

修改batch_size

mmdetection/configs/_base_/datasets/coco_detection.py

添加Tensorboard

mmdetection/configs/_base_/default_runtime.py

四、训练

使用Faster-RCNN训练数据

mmdetection使用配置文件的方式修改训练，直接在configs下复制faster-rcnn为my_faster-rcnn,在my_faster-rcnn进行修改，

1、修改基础配置文件

每个模型都有一个基础配置，其他规模的模型会基于该模型修改，这里只需要对基础模型配置文件修改即可，这里的基础模型是mmdetection/configs/_base_/models/fast-rcnn_r50_fpn.py，可以直接修改该文件也可以将该文件复制到my_faster-rcnn文件下，并修改对应配置文件中的 ../_base_/models/fast-rcnn_r50_fpn.py为 ./fast-rcnn_r50_fpn.py即可。

注：其他配置文件，如’coco_detection.py’, 'default_runtime.py’也可以使用复制到本地并修改的方式完成

如下图

2、训练

使用单GPU训练

python tools/train.py \
    ${CONFIG_FILE} \
    [optional arguments]
例子：
export CUDA_VISIBLE_DEVICES=0 python tools/train.py configs/mydino/dino-4scale_r50_8xb2-12e_coco.py 

在单机多GPU训练

bash ./tools/dist_train.sh \
    ${CONFIG_FILE} \
    ${GPU_NUM} \
    [optional arguments]
例子：
./tools/dist_train.sh configs/mask-rcnn_r50-caffe_fpn_ms-poly-1x_balloon.py 2

3、测试

为了测试训练完毕的模型，你只需要运行如下命令。

python tools/test.py configs/balloon/mask-rcnn_r50-caffe_fpn_ms-poly-1x_balloon.py work_dirs/mask-rcnn_r50-caffe_fpn_ms-poly-1x_balloon/epoch_12.pth

测试时输出标记后的图片

python tools/test.py configs/balloon/mask-rcnn_r50-caffe_fpn_ms-poly-1x_balloon.py work_dirs/mask-rcnn_r50-caffe_fpn_ms-poly-1x_balloon/epoch_12.pth --show