Spaces:
Sleeping
Sleeping
File size: 3,322 Bytes
8cc8f31 d54265f 8cc8f31 d54265f 8cc8f31 d54265f 8cc8f31 c9fce28 8cc8f31 d4b7ad4 8cc8f31 d54265f 8cc8f31 c9fce28 8cc8f31 c9fce28 8cc8f31 c9fce28 8cc8f31 |
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 |
from fsspec.parquet import open_parquet_file
import fsspec
import pyarrow.parquet as pq
from .grid import *
import pandas as pd
from io import BytesIO
import os
from PIL import Image
# GLOBAL VARIABLES
if os.path.isfile('helpers/s2l2a_metadata.parquet'):
l2a_meta_path = 'helpers/s2l2a_metadata.parquet'
else:
DATASET_NAME = 'Major-TOM/Core-S2L2A'
l2a_meta_path = 'https://huggingface.co/datasets/{}/resolve/main/metadata.parquet'.format(DATASET_NAME)
if os.path.isfile('helpers/s2l1c_metadata.parquet'):
l1c_meta_path = 'helpers/s2l1c_metadata.parquet'
else:
DATASET_NAME = 'Major-TOM/Core-S2L1C'
l1c_meta_path = 'https://huggingface.co/datasets/{}/resolve/main/metadata.parquet'.format(DATASET_NAME)
grid = Grid(10, latitude_range=(-90,90), longitude_range=(-180,180))
l2a_df = pd.read_parquet(l2a_meta_path)
l1c_df = pd.read_parquet(l1c_meta_path)
# HELPER FUNCTIONS
def gridcell2ints(grid_string):
up = int(grid_string.split('_')[0][:-1]) * (2*int(grid_string.split('_')[0][-1]=='U') - 1) # +ve if up
right = int(grid_string.split('_')[1][:-1]) * (2*int(grid_string.split('_')[1][-1]=='R') - 1) # +ve if R
return up, right
def row2image(parquet_url, parquet_row, fullrow_read=True):
if fullrow_read:
# option 1
f=fsspec.open(parquet_url)
temp_path = f.open()
else:
# option 2
temp_path = open_parquet_file(parquet_url,columns = ["thumbnail"])
with pq.ParquetFile(temp_path) as pf:
first_row_group = pf.read_row_group(parquet_row, columns=['thumbnail'])
stream = BytesIO(first_row_group['thumbnail'][0].as_py())
return Image.open(stream)
def row2s2(parquet_url, parquet_row, s2_bands = ["B04", "B03", "B02"]):
with open_parquet_file(parquet_url,columns = s2_bands) as f:
with pq.ParquetFile(f) as pf:
first_row_group = pf.read_row_group(parquet_row, columns=s2_bands)
return first_row_group
def cell2row(grid_string, meta_df, return_row = False):
row_U, col_R = gridcell2ints(grid_string)
R = meta_df.query('grid_row_u == {} & grid_col_r == {}'.format(row_U, col_R))
if not R.empty:
if return_row:
return R.parquet_url.item(), R.parquet_row.item(), R
else:
return R.parquet_url.item(), R.parquet_row.item()
else:
return None
def map_to_image(map, return_centre=False, return_gridcell=False, l2a=True):
try:
# 1. get bounds
bbox = map.get_bbox()
center = [(bbox[3]+bbox[1])/2, (bbox[2]+bbox[0])/2]
except:
return None
# 2. translate coordinate to major-tom tile
rows, cols = grid.latlon2rowcol([center[0]], [center[1]])
# 3. translate major-tom cell to row in parquet
df = l2a_df if l2a else l1c_df
row = cell2row("{}_{}".format(rows[0],cols[0]), df, return_row = True)
if row is not None:
parquet_url, parquet_row, meta_row = row
print(meta_row)
img = row2image(parquet_url, parquet_row)
# 4. acquire image # X. update map
lat, lon = meta_row.centre_lat.item(), meta_row.centre_lon.item()
ret = [img]
if return_centre:
ret.append((lat,lon))
if return_gridcell:
ret.append(meta_row.grid_cell.item())
return ret
else:
return None |