def load_file(pc3_path):
logging.info("Loading '" + pc3_path + "'")
cls_path = pc3_path[:-7] + 'CLS.txt'
pset = PointSet(pc3_path, cls_path)
psets = pset.split()
return [extract_data(ps) for ps in psets]
def pre_processor(files, input_queue):
for file in files:
logging.info('Loading {}'.format(file))
pset = PointSet(file)
psets = pset.split()
num_batches = int(math.ceil((1.0 * len(psets)) / BATCH_SIZE))
data = []
for batch_idx in range(num_batches):
start_idx = batch_idx * BATCH_SIZE
end_idx = (batch_idx + 1) * BATCH_SIZE
for k in range(FLAGS.n_angles):
batch_raw, batch_data = get_batch(psets, start_idx, end_idx)
if k == 0:
aug_data = batch_data
else:
ang = (1.0 * k) / (1.0 * FLAGS.n_angles) * 2 * np.pi
if FLAGS.extra_dims:
aug_data = np.concatenate(
(provider.rotate_point_cloud_z(
batch_data[:, :, 0:3],
angle=ang), batch_data[:, :, 3:]),
axis=2)
else:
aug_data = provider.rotate_point_cloud_z(batch_data)
data.append((batch_raw, aug_data))
logging.debug('Adding {} to queue'.format(file))
input_queue.put((pset, data))
logging.debug('Added {} to queue'.format(file))
logging.info('Pre-processing finished')
input_queue.put(None)
logging.debug('Pre-processing thread finished')
def post_processor(output_queue):
while True:
out_data = output_queue.get()
if out_data is None:
break
pset = out_data[0]
all_points = out_data[1]
all_labels = out_data[2]
logging.info('Post-processing {}'.format(pset.filename))
with tempfile.TemporaryDirectory() as tmpdir:
# Save pset to temp file
ipath = os.path.join(tmpdir, pset.filename + '_original.las')
pset.save(ipath)
# Update pset points
pset.x = all_points[:, 0]
pset.y = all_points[:, 1]
pset.z = all_points[:, 2]
pset.i = all_points[:, 3]
pset.r = all_points[:, 4]
pset.c = np.array([LABEL_MAP[v] for v in all_labels],
dtype='uint8')
# Save all classed points to a new file
cpath = os.path.join(tmpdir, pset.filename + '_candidates.las')
pset.save(cpath)
if FLAGS.output_type is OutputType.LABELS:
opath = os.path.join(tmpdir, pset.filename + '.las')
else:
opath = os.path.join(FLAGS.output_path, pset.filename + '.las')
# Run nearest neighbor voting algorithm to classify original points (pdal pipeline):
pipeline = {
'pipeline': [
ipath,
{
'type': 'filters.neighborclassifier',
'k': FLAGS.n_angles * 4 + 1,
'candidate': cpath
}, # Note: number of votes is FLAGS.n_angles*4+1, where 4 comes from splitting the point cloud (nominal number of overlapping subtiles per point before rotations)
opath
]
}
p = subprocess.run([
'/home/xaserver/anaconda3/envs/cpdal-run/bin/pdal', 'pipeline',
'-s'
],
input=json.dumps(pipeline).encode())
if p.returncode:
raise ValueError('Failed to run pipeline: \n"' +
json.dumps(pipeline) + '"')
if not FLAGS.output_type is OutputType.LAS:
# Load in updated point cloud, save classification file
pset2 = PointSet(opath)
pset2.save_classifications_txt(
os.path.join(FLAGS.output_path,
pset.filename + '_CLS.txt'))
output_queue.task_done()
logging.debug('Finished {}'.format(pset.filename))
logging.debug('Post-processing thread finished')
from random import randint from enum import Enum from math import pi, asin from gui import GUI from pointset import PointSet from algorithm import visibility_polygon pointset = PointSet(500, 500, visibility_polygon) gui = GUI(pointset) gui.mainloop()
from random import randint from enum import Enum from math import pi, asin from gui import GUI from pointset import PointSet from alg1 import convex_hull as algorithm1 from alg5 import convex_hull as algorithm5 pointset = PointSet(500, 500, algorithm5) gui = GUI(pointset) gui.mainloop()
from random import randint from enum import Enum from math import pi, asin from gui import GUI from pointset import PointSet from brute_circle_algorithm import enclosing_circle as brute_circle_algorithm from minidisc_algorithm import minidisc as minidisc_algorithm pointset = PointSet(500, 500, minidisc_algorithm) gui = GUI(pointset) gui.mainloop()
from gui import GUI from pointset import PointSet pointset = PointSet(500, 500) gui = GUI(pointset) gui.mainloop()
from random import randint from enum import Enum from math import pi, asin from gui import GUI from pointset import PointSet from smallest_rectangle import smallest_rectangle pointset = PointSet(500, 500, smallest_rectangle) gui = GUI(pointset) gui.mainloop()