-
Notifications
You must be signed in to change notification settings - Fork 4
/
StimulateExtreme.py
165 lines (135 loc) · 6.03 KB
/
StimulateExtreme.py
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
import xml.dom.minidom
import xml.etree.ElementTree as ET
import cv2
import os
from tqdm import tqdm
import albumentations as A
def prettyXml(element, indent, newline, level=0):
"""
Formats XML data to make it more readable.
"""
if element:
if element.text == None or element.text.isspace():
element.text = newline + indent * (level + 1)
else:
element.text = newline + indent * (level + 1) + element.text.strip() + newline + indent * (level + 1)
temp = list(element)
for subelement in temp:
if temp.index(subelement) < (len(temp) - 1):
subelement.tail = newline + indent * (level + 1)
else:
subelement.tail = newline + indent * level
prettyXml(subelement, indent, newline, level=level + 1)
return element
def read_xml(ann_path):
"""
Reads an XML file of an image and returns data in COCO dataset format.
"""
in_file = open(ann_path, "r", encoding="utf-8")
tree = ET.parse(in_file)
root = tree.getroot()
obj_dict = {
"image": None,
"bboxes": [],
"class_labels": []
}
for obj in root.iter('object'):
cls = obj.find('name').text
if cls == "waterweeds":
continue
xmlbox = obj.find('bndbox')
obj_dict["bboxes"].append([
float(xmlbox.find('xmin').text),
float(xmlbox.find('ymin').text),
float(xmlbox.find('xmax').text),
float(xmlbox.find('ymax').text)
])
obj_dict["class_labels"].append(cls)
return obj_dict
def save_xml(save_xml_path, image, bboxes, class_labels):
"""
Saves data to an XML file in a specified format.
"""
root = ET.Element("annotation")
folder = ET.SubElement(root, "folder")
folder.text = "images"
filename = ET.SubElement(root, "filename")
filename.text = "filename"
path = ET.SubElement(root, "path")
path.text = "path"
source = ET.SubElement(root, "source")
database = ET.SubElement(source, "database")
database.text = "Unknown"
size = ET.SubElement(root, "size")
width = ET.SubElement(size, "width")
width.text = str(image.shape[1])
height = ET.SubElement(size, "height")
height.text = str(image.shape[0])
depth = ET.SubElement(size, "depth")
depth.text = str(image.shape[-1])
segmented = ET.SubElement(root, "segmented")
segmented.text = str(0)
for i in range(len(bboxes)):
object = ET.SubElement(root, "object")
name = ET.SubElement(object, "name")
name.text = class_labels[i]
pose = ET.SubElement(object, "pose")
pose.text = "Unspecified"
truncated = ET.SubElement(object, "truncated")
truncated.text = str(0)
difficult = ET.SubElement(object, "difficult")
difficult.text = str(0)
bndbox = ET.SubElement(object, "bndbox")
xmin = ET.SubElement(bndbox, "xmin")
xmin.text = str(bboxes[i][0])
ymin = ET.SubElement(bndbox, "ymin")
ymin.text = str(bboxes[i][1])
xmax = ET.SubElement(bndbox, "xmax")
xmax.text = str(bboxes[i][2])
ymax = ET.SubElement(bndbox, "ymax")
ymax.text = str(bboxes[i][3])
root = prettyXml(root, '\t', '\n')
tree = ET.ElementTree(root)
tree.write(save_xml_path, encoding='utf-8')
def main():
num = 0
for img_name in tqdm(os.listdir(old_img_path)):
if img_name.split(".")[0] not in train_list:
continue
img_path = os.path.join(old_img_path, img_name)
if ".png" in img_name:
xml_name = img_name.replace(".png", ".xml")
elif ".jpg" in img_name:
xml_name = img_name.replace(".jpg", ".xml")
else:
print("Unsupported file type:", img_name)
exit()
xml_path = os.path.join(old_xml_path, xml_name)
ann = read_xml(xml_path)
img = cv2.imread(img_path)
ann["image"] = img
print(xml_path)
for i in range(len(transform_list)):
aug_name = str(transform_list[i][0]).split("(")[0]
ann_aug = transform_list[i](image=ann["image"], bboxes=ann["bboxes"], class_labels=ann["class_labels"])
ann_aug["bboxes"] = [list(map(int, list(bbox))) for bbox in ann_aug["bboxes"]]
save_img_name = "Aug_id_" + str(i) + "_" + aug_name + "_" + img_name
save_xml_name = "Aug_id_" + str(i) + "_" + aug_name + "_" + xml_name
save_img_path = os.path.join(new_img_file, save_img_name)
save_xml_path = os.path.join(new_aug_file, save_xml_name)
cv2.imwrite(save_img_path, ann_aug["image"])
save_xml(save_xml_path, ann_aug["image"], ann_aug["bboxes"], ann_aug["class_labels"])
# print(save_img_path)
if __name__ == "__main__":
# Albumentations documentation: https://albumentations.ai/docs/
# How to use Albumentations: https://github.com/zk2ly/How-to-use-Albumentations
classes = ["robot"]
transform_list = [
A.Compose(A.RandomRain(brightness_coefficient=0.9, drop_width=1, blur_value=5, p=1), bbox_params=A.BboxParams(format='pascal_voc', min_area=0., min_visibility=0., label_fields=['class_labels'])),
A.Compose(A.RandomFog(0.6, p=1), bbox_params=A.BboxParams(format='pascal_voc', min_area=0., min_visibility=0., label_fields=['class_labels'])),
A.Compose(A.MotionBlur(p=1), bbox_params=A.BboxParams(format='pascal_voc', min_area=0., min_visibility=0., label_fields=['class_labels'])),
]
old_xml_path = r"D:\ProgrammingProjects\PythonProjects\Contextual-Object-Detection\CCTSDB_VOC\VOC2007\Annotations"
old_img_path = r"D:\ProgrammingProjects\PythonProjects\Contextual-Object-Detection\CCTSDB_VOC\VOC2007\JPEGImages"
new_aug_file = r"D:\ProgrammingProjects\PythonProjects\Contextual-Object-Detection\CCTSDB_VOC_aug\VOC2007\Annotations"
new_img_file = r"D:\ProgrammingProjects\PythonProjects\Context