40 lines
1.3 KiB
Python
40 lines
1.3 KiB
Python
import xmltodict
|
|
import numpy as np
|
|
import glob
|
|
import cv2
|
|
|
|
def xml_to_dataset(dir, size=None):
|
|
tab_image=[]
|
|
tab_label=[]
|
|
for fichier in glob.glob(dir+"/*.xml"):
|
|
with open(fichier) as fd:
|
|
doc=xmltodict.parse(fd.read())
|
|
image=doc['annotation']['filename']
|
|
img=cv2.imread(dir+"/"+image)
|
|
objects=doc['annotation']['object'] if type(doc['annotation']['object'])==list else [doc['annotation']['object']]
|
|
for obj in objects:
|
|
xmin=int(obj['bndbox']['xmin'])
|
|
xmax=int(obj['bndbox']['xmax'])
|
|
ymin=int(obj['bndbox']['ymin'])
|
|
ymax=int(obj['bndbox']['ymax'])
|
|
if size is not None:
|
|
tab_image.append(cv2.resize(img[ymin:ymax, xmin:xmax], size))
|
|
else:
|
|
tab_image.append(img[ymin:ymax, xmin:xmax])
|
|
tab_label.append(obj['name'])
|
|
|
|
l=list(set(tab_label))
|
|
tab_one_hot=[]
|
|
for e in tab_label:
|
|
tab_one_hot.append(np.eye(len(l))[l.index(e)])
|
|
|
|
return tab_image, l, tab_one_hot
|
|
|
|
tab_image, tab_label, tab_one_hot=xml_to_dataset("./", (32, 32))
|
|
|
|
for i in range(len(tab_image)):
|
|
cv2.imshow('image', tab_image[i])
|
|
print(tab_label[np.argmax(tab_one_hot[i])], tab_one_hot[i])
|
|
cv2.waitKey()
|
|
|