rasha-salim

import rasterio
from rasterio import features
from rasterio.io import MemoryFile
import math
import geopandas as gpd
import os 


w = 400
h = 400

rasha-salim

def getGeojson():
    # Store the mask coordinates into a geojson file
    bounds = getImgBounds()
    dataset = rasterio.open(png_path)          
    bands = [1]
    data = dataset.read(bands)
    transform = rasterio.transform.from_bounds(bounds[0], bounds[1], bounds[2], bounds[3], data.shape[1], data.shape[2])
    crs = rasterio.crs.CRS({"init": "epsg:4326"})   # 3857 Google Maps Projection    4326  World wide (3D)
    with MemoryFile() as memfile:
        meta = {"count": 1, "width": data.shape[1], "height": data.shape[2], "transform": transform, "nodata": 0, "crs": crs, "dtype":data.dtype}

rasha-salim

def getGeojson():
    # Store the mask coordinates into a geojson file
    bounds = getImgBounds()
    dataset = rasterio.open(png_path)          
    bands = [1]
    data = dataset.read(bands)
    transform = rasterio.transform.from_bounds(bounds[0], bounds[1], bounds[2], bounds[3], data.shape[1], data.shape[2])
    crs = rasterio.crs.CRS({"init": "epsg:4326"})   # 3857 Google Maps Projection    4326  World wide (3D)
    with MemoryFile() as memfile:
        meta = {"count": 1, "width": data.shape[1], "height": data.shape[2], "transform": transform, "nodata": 0, "crs": crs, "dtype":data.dtype}

rasha-salim

import os 
import math
import rasterio
from rasterio import features

# Specify the image parameters
w = 400
h = 400
zoom = 8
lat = -19.361500892883598

rasha-salim

# I just extracted the input, rgb, and segmentation to separate arrays for convenience
inputs = []
masks = []
rgb_inputs = []
for pairs in images_target:
  inputs.append(pairs['input'])
  masks.append(pairs['mask'])
  rgb_inputs.append(pairs['input_rgb'])
  
# Pass the data to our Dataset instance we created earlier

rasha-salim

inputs_dir = '/content/landcovernet/inputs'  # The directory where our input resides
targets_dir = '/content/landcovernet/targets' # The directory where our target resides
# First stack all the bands togather

def process_tiffs(inputs_dir, target_dir):
  data = []
  sub_dir_list = []
  images_target = {}
  stacked_imgs = []
  list_bands = []

rasha-salim

inputs_dir = '/content/landcovernet/inputs'  # The directory where our input resides
targets_dir = '/content/landcovernet/targets' # The directory where our target resides
# First stack all the bands togather

def process_tiffs(inputs_dir, target_dir):
  data = []
  sub_dir_list = []
  images_target = {}
  stacked_imgs = []
  list_bands = []

rasha-salim

inputs_dir = '/content/landcovernet/inputs'  # The directory where our input resides
targets_dir = '/content/landcovernet/targets' # The directory where our target resides
# First stack all the bands togather

def process_tiffs(inputs_dir, target_dir):
  data = []
  sub_dir_list = []
  images_target = {}
  stacked_imgs = []
  list_bands = []

rasha-salim

import hub
from hub.schema import Image
from hub.schema import Tensor
from hub.schema import Mask
from hub.schema import Segmentation

# include your user anme and a name for your dataset
tag = "rasha/landCoverNet_Omdena_Sample"
ds = {}
# Define youe dataset object

rasha-salim