Spaces:
Sleeping
Sleeping
File size: 7,454 Bytes
9bf4bd7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 |
# Copyright (c) OpenMMLab. All rights reserved.
import argparse
import math
import os
import os.path as osp
import mmcv
import mmengine
from mmocr.utils import crop_img, dump_ocr_data
def collect_files(img_dir, gt_dir, ratio):
"""Collect all images and their corresponding groundtruth files.
Args:
img_dir (str): The image directory
gt_dir (str): The groundtruth directory
ratio (float): Split ratio for val set
Returns:
files (list): The list of tuples (img_file, groundtruth_file)
"""
assert isinstance(img_dir, str)
assert img_dir
assert isinstance(gt_dir, str)
assert gt_dir
assert isinstance(ratio, float)
assert ratio < 1.0, 'val_ratio should be a float between 0.0 to 1.0'
ann_list, imgs_list = [], []
for ann_file in os.listdir(gt_dir):
ann_list.append(osp.join(gt_dir, ann_file))
imgs_list.append(osp.join(img_dir, ann_file.replace('txt', 'jpg')))
all_files = list(zip(imgs_list, ann_list))
assert len(all_files), f'No images found in {img_dir}'
print(f'Loaded {len(all_files)} images from {img_dir}')
trn_files, val_files = [], []
if ratio > 0:
for i, file in enumerate(all_files):
if i % math.floor(1 / ratio):
trn_files.append(file)
else:
val_files.append(file)
else:
trn_files, val_files = all_files, []
print(f'training #{len(trn_files)}, val #{len(val_files)}')
return trn_files, val_files
def collect_annotations(files, nproc=1):
"""Collect the annotation information.
Args:
files (list): The list of tuples (image_file, groundtruth_file)
nproc (int): The number of process to collect annotations
Returns:
images (list): The list of image information dicts
"""
assert isinstance(files, list)
assert isinstance(nproc, int)
if nproc > 1:
images = mmengine.track_parallel_progress(
load_img_info, files, nproc=nproc)
else:
images = mmengine.track_progress(load_img_info, files)
return images
def load_img_info(files):
"""Load the information of one image.
Args:
files (tuple): The tuple of (img_file, groundtruth_file)
Returns:
img_info (dict): The dict of the img and annotation information
"""
assert isinstance(files, tuple)
img_file, gt_file = files
assert osp.basename(gt_file).split('.')[0] == osp.basename(img_file).split(
'.')[0]
# read imgs while ignoring orientations
img = mmcv.imread(img_file)
img_info = dict(
file_name=osp.join(osp.basename(img_file)),
height=img.shape[0],
width=img.shape[1],
segm_file=osp.join(osp.basename(gt_file)))
if osp.splitext(gt_file)[1] == '.txt':
img_info = load_txt_info(gt_file, img_info)
else:
raise NotImplementedError
return img_info
def load_txt_info(gt_file, img_info):
"""Collect the annotation information.
The annotation format is as the following:
x1, y1, x2, y2, x3, y3, x4, y4, difficult, text
390,902,1856,902,1856,1225,390,1225,0,"金氏眼镜"
1875,1170,2149,1170,2149,1245,1875,1245,0,"创于1989"
2054,1277,2190,1277,2190,1323,2054,1323,0,"城建店"
Args:
gt_file (str): The path to ground-truth
img_info (dict): The dict of the img and annotation information
Returns:
img_info (dict): The dict of the img and annotation information
"""
anno_info = []
with open(gt_file, encoding='utf-8-sig') as f:
lines = f.readlines()
for line in lines:
points = line.split(',')[0:8]
word = line.split(',')[9].rstrip('\n').strip('"')
difficult = 1 if line.split(',')[8] != '0' else 0
bbox = [int(pt) for pt in points]
if word == '###' or difficult == 1:
continue
anno = dict(bbox=bbox, word=word)
anno_info.append(anno)
img_info.update(anno_info=anno_info)
return img_info
def generate_ann(root_path, split, image_infos, preserve_vertical):
"""Generate cropped annotations and label txt file.
Args:
root_path (str): The root path of the dataset
split (str): The split of dataset. Namely: training or val
image_infos (list[dict]): A list of dicts of the img and
annotation information
preserve_vertical (bool): Whether to preserve vertical texts
"""
dst_image_root = osp.join(root_path, 'crops', split)
ignore_image_root = osp.join(root_path, 'ignores', split)
if split == 'training':
dst_label_file = osp.join(root_path, 'train_label.json')
elif split == 'val':
dst_label_file = osp.join(root_path, 'val_label.json')
mmengine.mkdir_or_exist(dst_image_root)
mmengine.mkdir_or_exist(ignore_image_root)
img_info = []
for image_info in image_infos:
index = 1
src_img_path = osp.join(root_path, 'imgs', image_info['file_name'])
image = mmcv.imread(src_img_path)
src_img_root = image_info['file_name'].split('.')[0]
for anno in image_info['anno_info']:
word = anno['word']
dst_img = crop_img(image, anno['bbox'], 0, 0)
h, w, _ = dst_img.shape
dst_img_name = f'{src_img_root}_{index}.png'
index += 1
# Skip invalid annotations
if min(dst_img.shape) == 0:
continue
# Filter out vertical texts
if not preserve_vertical and h / w > 2:
dst_img_path = osp.join(ignore_image_root, dst_img_name)
mmcv.imwrite(dst_img, dst_img_path)
continue
dst_img_path = osp.join(dst_image_root, dst_img_name)
mmcv.imwrite(dst_img, dst_img_path)
img_info.append({
'file_name': dst_img_name,
'anno_info': [{
'text': word
}]
})
dump_ocr_data(img_info, dst_label_file, 'textrecog')
def parse_args():
parser = argparse.ArgumentParser(
description='Generate training and val set of RCTW.')
parser.add_argument('root_path', help='Root dir path of RCTW')
parser.add_argument(
'--val-ratio', help='Split ratio for val set', default=0.0, type=float)
parser.add_argument(
'--nproc', default=1, type=int, help='Number of process')
parser.add_argument(
'--preserve-vertical',
help='Preserve samples containing vertical texts',
action='store_true')
args = parser.parse_args()
return args
def main():
args = parse_args()
root_path = args.root_path
ratio = args.val_ratio
trn_files, val_files = collect_files(
osp.join(root_path, 'imgs'), osp.join(root_path, 'annotations'), ratio)
# Train set
with mmengine.Timer(
print_tmpl='It takes {}s to convert RCTW Training annotation'):
trn_infos = collect_annotations(trn_files, nproc=args.nproc)
generate_ann(root_path, 'training', trn_infos, args.preserve_vertical)
# Val set
if len(val_files) > 0:
with mmengine.Timer(
print_tmpl='It takes {}s to convert RCTW Val annotation'):
val_infos = collect_annotations(val_files, nproc=args.nproc)
generate_ann(root_path, 'val', val_infos, args.preserve_vertical)
if __name__ == '__main__':
main()
|