label_save_path = '/data/gaofen/gaofen3/v1/{}/labels'.format(image_set)
wwtool.mkdir_or_exist(label_save_path)
# print(os.listdir(label_path))
for idx, label_file in enumerate(os.listdir(label_path)):
print(idx, label_file)
file_name = label_file.split('.xml')[0]
label_file = os.path.join(label_path, file_name + '.xml')
image_file = os.path.join(image_path, file_name + '.tif')
img = imread(image_file)
objects = wwtool.rovoc_parse(label_file)
bboxes = np.array([obj['bbox'] for obj in objects])
subimages = wwtool.split_image(img, subsize=1024, gap=200)
subimage_coordinates = list(subimages.keys())
bboxes_ = bboxes.copy()
for subimage_coordinate in subimage_coordinates:
objects = []
bboxes_[:, 0] = bboxes[:, 0] - subimage_coordinate[0]
bboxes_[:, 1] = bboxes[:, 1] - subimage_coordinate[1]
cx_bool = np.logical_and(bboxes_[:, 0] >= 0,
bboxes_[:, 0] < 1024)
cy_bool = np.logical_and(bboxes_[:, 1] >= 0,
bboxes_[:, 1] < 1024)
subimage_bboxes = bboxes_[np.logical_and(cx_bool, cy_bool)]
if len(subimage_bboxes) == 0:
def split_image(self, image_fn):
if not image_fn.endswith('.jpg'):
return
image_file = os.path.join(self.image_path, image_fn)
shp_file = os.path.join(self.merged_shp_path, image_fn.replace('jpg', 'shp'))
geo_file = os.path.join(self.geo_path, image_fn.replace('jpg', 'png'))
file_name = os.path.splitext(os.path.basename(image_file))[0]
if not os.path.exists(shp_file):
return
img = cv2.imread(image_file)
geo_info = rio.open(geo_file)
objects = self.shp_parser(shp_file, geo_info)
masks = np.array([obj['segmentation'] for obj in objects])
subimages = wwtool.split_image(img, subsize=self.subimage_size, gap=self.gap)
subimage_coordinates = list(subimages.keys())
if masks.shape[0] == 0:
return
mask_centroids = []
for obj in objects:
geometry = obj['converted_polygon'].centroid
geo = geojson.Feature(geometry=geometry, properties={})
coordinate = geo.geometry["coordinates"]
coordinate[0], coordinate[1] = abs(coordinate[0]), abs(coordinate[1])
mask_centroids.append(coordinate)
mask_centroids = np.array(mask_centroids)
mask_centroids_ = mask_centroids.copy()
for subimage_coordinate in subimage_coordinates:
objects = []
mask_centroids_[:, 0] = mask_centroids[:, 0] - subimage_coordinate[0]
mask_centroids_[:, 1] = mask_centroids[:, 1] - subimage_coordinate[1]
cx_bool = np.logical_and(mask_centroids_[:, 0] >= 0, mask_centroids_[:, 0] < subimage_size)
cy_bool = np.logical_and(mask_centroids_[:, 1] >= 0, mask_centroids_[:, 1] < subimage_size)
subimage_masks = masks[np.logical_and(cx_bool, cy_bool)]
subimage_masks_ = []
for subimage_mask in subimage_masks:
subimage_mask_np = np.array(subimage_mask)
subimage_mask_np[0::2] = subimage_mask_np[0::2] - subimage_coordinate[0]
subimage_mask_np[1::2] = subimage_mask_np[1::2] - subimage_coordinate[1]
subimage_masks_.append(subimage_mask_np.tolist())
subimage_masks = subimage_masks_
if len(subimage_masks) == 0:
continue
img = subimages[subimage_coordinate]
if np.mean(img) == 0:
continue
label_save_file = os.path.join(self.label_save_path, '{}__{}_{}.txt'.format(file_name, subimage_coordinate[0], subimage_coordinate[1]))
image_save_file = os.path.join(self.image_save_path, '{}__{}_{}.png'.format(file_name, subimage_coordinate[0], subimage_coordinate[1]))
cv2.imwrite(image_save_file, img)
for subimage_mask in subimage_masks:
subimage_objects = dict()
subimage_objects['mask'] = subimage_mask
subimage_objects['label'] = 'building'
objects.append(subimage_objects)
wwtool.simpletxt_dump(objects, label_save_file, encode='mask')
def split_image(self, image_fn):
if not image_fn.endswith('.jpg'):
return
image_file = os.path.join(self.image_path, image_fn)
shp_file = os.path.join(self.roof_shp_path,
image_fn.replace('jpg', 'shp'))
geo_file = image_file
file_name = os.path.splitext(os.path.basename(image_file))[0]
if not os.path.exists(shp_file):
return
img = cv2.imread(image_file)
geo_info = rio.open(geo_file)
objects = self.shp_parser(shp_file,
geo_info,
coord='pixel',
ignore_file=None,
merge_flag=True,
connection_mode='floor')
masks = np.array([obj['segmentation'] for obj in objects])
subimages = wwtool.split_image(img,
subsize=self.subimage_size,
gap=self.gap)
subimage_coordinates = list(subimages.keys())
if masks.shape[0] == 0:
return
mask_centroids = []
for obj in objects:
geometry = obj['converted_polygon'].centroid
geo = geojson.Feature(geometry=geometry, properties={})
coordinate = geo.geometry["coordinates"]
coordinate[0], coordinate[1] = abs(coordinate[0]), abs(
coordinate[1])
mask_centroids.append(coordinate)
mask_centroids = np.array(mask_centroids)
mask_centroids_ = mask_centroids.copy()
for subimage_coordinate in subimage_coordinates:
objects = []
mask_centroids_[:,
0] = mask_centroids[:, 0] - subimage_coordinate[0]
mask_centroids_[:,
1] = mask_centroids[:, 1] - subimage_coordinate[1]
cx_bool = np.logical_and(mask_centroids_[:, 0] >= 0,
mask_centroids_[:, 0] < subimage_size)
cy_bool = np.logical_and(mask_centroids_[:, 1] >= 0,
mask_centroids_[:, 1] < subimage_size)
subimage_masks = masks[np.logical_and(cx_bool, cy_bool)]
subimage_masks_ = []
for subimage_mask in subimage_masks:
if wwtool.mask2polygon(subimage_mask).area < 5:
continue
subimage_mask_np = np.array(subimage_mask)
subimage_mask_np[
0::2] = subimage_mask_np[0::2] - subimage_coordinate[0]
subimage_mask_np[
1::2] = subimage_mask_np[1::2] - subimage_coordinate[1]
subimage_masks_.append(subimage_mask_np.tolist())
subimage_masks = subimage_masks_
# cut the polygons by image boundary
subimage_masks = wwtool.clip_mask(subimage_masks,
image_size=(1024, 1024))
# judge whether to drop this subimage
drop_flag = self.drop_subimage(subimages,
subimage_coordinate,
subimage_masks,
show=self.show)
if drop_flag:
continue
img = subimages[subimage_coordinate]
label_save_file = os.path.join(
self.label_save_path,
'{}_{}__{}_{}.txt'.format(self.sub_imageset_fold, file_name,
subimage_coordinate[0],
subimage_coordinate[1]))
image_save_file = os.path.join(
self.image_save_path,
'{}_{}__{}_{}.png'.format(self.sub_imageset_fold, file_name,
subimage_coordinate[0],
subimage_coordinate[1]))
cv2.imwrite(image_save_file, img)
for subimage_mask in subimage_masks:
subimage_objects = dict()
subimage_objects['mask'] = subimage_mask
subimage_objects['label'] = 'building'
objects.append(subimage_objects)
wwtool.simpletxt_dump(objects, label_save_file, encode='mask')
frame_id += 1
continue
print(scene_name, video_name, frame_id, image_name)
objects = stanford_compus_parse.stanford_compus_parse(
scene_name, video_name, frame_id)
if objects == []:
frame_id += 1
continue
bboxes = np.array(
[wwtool.xyxy2cxcywh(obj['bbox']) for obj in objects])
labels = np.array([obj['label'] for obj in objects])
subimages = wwtool.split_image(img,
subsize=subimage_size,
gap=gap,
expand_boundary=True)
subimage_coordinates = list(subimages.keys())
bboxes_ = bboxes.copy()
labels_ = labels.copy()
if bboxes_.shape[0] == 0:
frame_id += 1
continue
for subimage_coordinate in subimage_coordinates:
objects = []
bboxes_[:, 0] = bboxes[:, 0] - subimage_coordinate[0]
bboxes_[:, 1] = bboxes[:, 1] - subimage_coordinate[1]
cx_bool = np.logical_and(bboxes_[:, 0] >= 0,
bboxes_[:, 0] < subimage_size)
# print(os.listdir(label_path))
for idx, label_file in enumerate(os.listdir(label_path)):
print(idx, label_file)
file_name = label_file.split('.txt')[0]
label_file = os.path.join(label_path, file_name + '.txt')
image_file = os.path.join(image_path, file_name + '.jpg')
img = imread(image_file)
objects = wwtool.nwpu_parse(label_file)
bboxes = np.array(
[wwtool.xyxy2cxcywh(obj['bbox']) for obj in objects])
labels = np.array([obj['label'] for obj in objects])
subimages = wwtool.split_image(img, subsize=subimage_size, gap=gap)
subimage_coordinates = list(subimages.keys())
bboxes_ = bboxes.copy()
labels_ = labels.copy()
if bboxes_.shape[0] == 0:
continue
for subimage_coordinate in subimage_coordinates:
objects = []
bboxes_[:, 0] = bboxes[:, 0] - subimage_coordinate[0]
bboxes_[:, 1] = bboxes[:, 1] - subimage_coordinate[1]
cx_bool = np.logical_and(bboxes_[:, 0] >= 0,
bboxes_[:, 0] < subimage_size)
cy_bool = np.logical_and(bboxes_[:, 1] >= 0,
from wwtool import split_image, show_image
from skimage.io import imread
if __name__ == '__main__':
# img = np.random.rand(5292, 3371)
img_path = '/home/jwwangchn/data/GF/v0/tif'
save_path = '/home/jwwangchn/data/GF/v1/vis'
for image_name in os.listdir(img_path):
mmcv.mkdir_or_exist(
os.path.join(save_path,
image_name.split('.png')[0]))
# img = imread(os.path.join(img_path, image_name))
img = cv2.imread(os.path.join(img_path, image_name))
print("{} has the shape {}".format(image_name, img.shape))
subimages = split_image(img, 1024, 0, mode='keep_all')
subimage_coordinates = list(subimages.keys())
print("start point: ", subimage_coordinates)
for subimage_coordinate in subimage_coordinates:
print("coordinate: {}".format(subimage_coordinate))
img = subimages[subimage_coordinate]
if np.mean(img) == 0:
continue
save_name = os.path.join(
save_path,
image_name.split('.png')[0], "{}__{}_{}.png".format(
image_name.split('.png')[0], subimage_coordinate[0],
subimage_coordinate[1]))
cv2.imwrite(save_name, img)
# show_image(img, wait_time=100, save_name=save_name)
import os
import cv2
import numpy as np
import mmcv
from wwtool import split_image, show_image
from skimage.io import imread
if __name__ == '__main__':
# img = np.random.rand(5292, 3371)
img_path = '/data/gaofen/v0/tif'
save_path = '/data/gaofen/v1/vis'
for image_name in os.listdir(img_path):
mmcv.mkdir_or_exist(os.path.join(save_path, image_name.split('.tif')[0]))
img = imread(os.path.join(img_path, image_name))
print("{} has the shape {}".format(image_name, img.shape))
subimages = split_image(img, np.minimum(img.shape[0], img.shape[1])//2, 0, mode='drop_boundary')
subimage_coordinates = list(subimages.keys())
print("start point: ", subimage_coordinates)
for subimage_coordinate in subimage_coordinates:
print("coordinate: {}".format(subimage_coordinate))
img = subimages[subimage_coordinate]
if np.mean(img) == 0:
continue
save_name = os.path.join(save_path, image_name.split('.tif')[0], "{}__{}_{}.png".format(image_name.split('.tif')[0], subimage_coordinate[0], subimage_coordinate[1]))
cv2.imwrite(save_name, img)
# show_image(img, wait_time=100, save_name=save_name)
