Ultra-Fast-Lane-Detection 制作自己的数据集并进行训练

几个不错的博客：

【Lane】 Ultra-Fast-Lane-Detection 复现 by 摇曳的树

车道线检测模型笔记 by 小驴淘米666

ultra fast lane detection数据集制作 by 小王在线秃头

（1）使用labelme进行车道线标注

labelme打开需要标注的图片文件夹./datasets/imgs，选择Creat LineStrip进行车道线标注，并将标注文件存至./datasets/jsons

（2）labelme_json_to_dataset批量进行标注文件转换

在datasets目录下运行 python labelme_json_to_dataset.py ./jsons 

import argparse 
import json 
import os
import os.path as osp 
import warnings 
import PIL.Image 
import yaml
 
from labelme import utils 
import base64
 
#批量转换代码
def main():
    warnings.warn("This script is aimed to demonstrate how to convert the\n"
                  "JSON file to a single image dataset, and not to handle\n"
                  "multiple JSON files to generate a real-use dataset.")
    parser = argparse.ArgumentParser()
    parser.add_argument('json_file')
    parser.add_argument('-o', '--out', default=None) 
    args = parser.parse_args()
    json_file = args.json_file

    if args.out is None:
        out_dir = osp.basename(json_file).replace('.', '_')
        out_dir = osp.join(osp.dirname(json_file), out_dir)
    else:
        out_dir = args.out
 
    if not osp.exists(out_dir):
        os.mkdir(out_dir)
 
    count = os.listdir(json_file)
 
    for i in range(0, len(count)):
        path = os.path.join(json_file, count[i])
        print(path)
 
        if os.path.isfile(path):
            data = json.load(open(path, encoding='UTF8'))

            if data['imageData']:
                imageData = data['imageData']
            else:
                imagePath = os.path.join(os.path.dirname(path), data['imagePath'])
                with open(imagePath, 'rb') as f:
                    imageData = f.read()
                    imageData = base64.b64encode(imageData).decode('utf-8')

            img = utils.img_b64_to_arr(imageData)
            label_name_to_value = {'_background_': 0}
 
            for shape in data['shapes']:
                label_name = shape['label']

                if label_name in label_name_to_value:
 
                    label_value = label_name_to_value[label_name]
                else:
                    label_value = len(label_name_to_value)
                    label_name_to_value[label_name] = label_value
 
            # label_values must be dense
            label_values, label_names = [], []
            for ln, lv in sorted(label_name_to_value.items(), key=lambda x: x[1]):
                label_values.append(lv) 
                label_names.append(ln)
 
            assert label_values == list(range(len(label_values)))
 
            lbl = utils.shapes_to_label(img.shape, data['shapes'], label_name_to_value) 
            captions = ['{}: {}'.format(lv, ln)
                for ln, lv in label_name_to_value.items()]
 
            lbl_viz = utils.draw_label(lbl, img, captions)
            out_dir = osp.basename(count[i]).replace('.', '_')
            out_dir = osp.join(osp.dirname(count[i]), out_dir)

            if not osp.exists(out_dir):
                os.mkdir(out_dir)
 
            PIL.Image.fromarray(img).save(osp.join(out_dir, 'img.png'))
            #PIL.Image.fromarray(lbl).save(osp.join(out_dir, 'label.png'))
            utils.lblsave(osp.join(out_dir, 'label.png'), lbl)
            PIL.Image.fromarray(lbl_viz).save(osp.join(out_dir, 'label_viz.png'))
 
            with open(osp.join(out_dir, 'label_names.txt'), 'w') as f: 
                for lbl_name in label_names:
                    f.write(lbl_name + '\n')
 
            warnings.warn('info.yaml is being replaced by label_names.txt')
            info = dict(label_names=label_names)

            with open(osp.join(out_dir, 'info.yaml'), 'w') as f:
                yaml.safe_dump(info, f, default_flow_style=False)
 
            print('Saved to: %s' % out_dir)
 
if __name__ == '__main__':
    main()

生成的每个文件夹中包含下列5个文件

 将文件统一移至./datasets/annotations

（3）生成训练文件train_data

运行./datasets/gen_train_gt.py

import cv2
from skimage import measure, color
from skimage.measure import regionprops
import numpy as np
import os
import copy
from PIL import Image



def skimageFilter(gray):
    binary_warped = copy.copy(gray)
    binary_warped[binary_warped > 0.1] = 255

    gray = (np.dstack((gray, gray, gray)) * 255).astype('uint8')
    labels = measure.label(gray[:, :, 0], connectivity=1)
    dst = color.label2rgb(labels, bg_label=0, bg_color=(0, 0, 0))
    gray = cv2.cvtColor(np.uint8(dst * 255), cv2.COLOR_RGB2GRAY)
    return binary_warped, gray


def moveImageTodir(path, targetPath, name):
    if os.path.isdir(path):
        image_name = "gt_image/" + str(name) + ".png"
        binary_name = "gt_binary_image/" + str(name) + ".png"
        instance_name = "gt_instance_image/" + str(name) + ".png"

        # train_rows = image_name + " " + binary_name + " " + instance_name + "\n"
        train_rows = image_name  + " " + instance_name + "\n"

        origin_img = cv2.imread(path + "/img.png")
        origin_img = cv2.resize(origin_img, (1280, 720))
        cv2.imwrite(targetPath + "/" + image_name, origin_img)
        print(targetPath + "/" + image_name)

        # img = cv2.imread(path + '/label.png')
        # img = cv2.resize(img, (1280, 720))
        # gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) #将bgr格式的图片转换成灰度图片
        # binary_warped, instance = skimageFilter(gray)
        # cv2.imwrite(targetPath + "/" + binary_name, binary_warped)
        # print(targetPath + "/" + binary_name)
        # cv2.imwrite(targetPath + "/" + instance_name, instance)

        ins = Image.open(path + '/label.png')
        ins = ins.resize((1280,720))
        ins.save(targetPath + "/" + instance_name)

        print("success create data name is : ", train_rows)
        return train_rows
    return None


if __name__ == "__main__":

    count = 1
    with open(r"/media/ai/D/Teamwork/wushuli/LaneDet/Ultra-Fast-Lane-Detection-master/datasets/train_data/train_gt.txt", 'w+') as file:
        dir_name = r"/media/ai/D/Teamwork/wushuli/LaneDet/Ultra-Fast-Lane-Detection-master/datasets/annotations"
        for annotations_dir in os.listdir(dir_name):
            json_dir = os.path.join(dir_name, annotations_dir)
            # print(json_dir)
            target_path = r"/media/ai/D/Teamwork/wushuli/LaneDet/Ultra-Fast-Lane-Detection-master/datasets/train_data"
            if os.path.isdir(json_dir):
                train_rows = moveImageTodir(json_dir,target_path, str(count).zfill(4))
                file.write(train_rows)
                count += 1

整体目录结构

 （4）进行训练

修改配置文件参数configs/tusimple.py

data_root：训练数据集train_data路径
log_path：训练记录和模型存储位置

运行 python train.py configs/tusimple.py进行训练