mlubej

import cv2
import numpy as np


def get_hillshade(dem: np.ndarray, azimuth: float, zenith: float, resolution: int):
    """Calculates hillshade from DEM."""

    # calculate gradient using open-cv
    grad_x = cv2.Sobel(dem, cv2.CV_32F, 1, 0, ksize=3) / resolution / 8
    grad_y = cv2.Sobel(dem, cv2.CV_32F, 0, 1, ksize=3) / resolution / 8

mlubej

import json
import subprocess
import sys
from typing import Dict, List

import numpy as np


def get_nth_space(index: int) -> int:
    spaces = get_non_native_spaces()

mlubej

//VERSION=3

function setup() {
  return {
    input: ["B04", "B08", "CLP", "dataMask"],
    output: {bands: 2, sampleType: SampleType.INT8},
    mosaicking: "ORBIT"
  }
}

mlubej

import subprocess
from sentinelhub import CRS

# rescale to a lower resolution
nres = 120*10
target_crs = CRS.WGS84.epsg

def create_rgb_per_timestamp(batch_tile_name):
    subprocess.run(f'gdalwarp -tr {nres} {nres} <INPUT HI-RES RED> <OUTPUT LO_RES RED>', shell=True)
    # repeat for green and blue ...

mlubej

import subprocess

# tasks for exporting to .tiffs
export_r = ExportToTiff(feature=(FeatureType.DATA, 'RGB'), folder='./tiffs/', band_indices=[0])
export_g = ExportToTiff(feature=(FeatureType.DATA, 'RGB'), folder='./tiffs/', band_indices=[1])
export_b = ExportToTiff(feature=(FeatureType.DATA, 'RGB'), folder='./tiffs/', band_indices=[2])

# gdal magic for spatial merge
subprocess.run(f'gdal_merge.py -n 0 -a_nodata 0 -o tiffs/r.tiff -co compress=LZW tiffs/r_patch_*.tiff && rm -rf tiffs/r_patch_*.tiff', shell=True);
subprocess.run(f'gdal_merge.py -n 0 -a_nodata 0 -o tiffs/g.tiff -co compress=LZW tiffs/g_patch_*.tiff && rm -rf tiffs/g_patch_*.tiff', shell=True);

mlubej

import ffmpeg

# video related
stream = ffmpeg.input('<INPUT_IMAGES>', pattern_type='glob', framerate=<FPS>)  # load image stream
stream = stream.filter('pad', w='ceil(iw/2)*2', h='ceil(ih/2)*2', color='white')  # pad the image
split = stream.split()  # split for multiple usecases
video = split[0]  # define the video stream

# gif related
palette = split[1].filter('palettegen', reserve_transparent=True, stats_mode='diff')  # determine time-lapse specific color pallete to reduce dithering

mlubej

import ephem
import numpy as np
from datetime import datetime

def get_shadow_length(long, lat, datetime, height):
    """
    Accepts longitude and latitude in the WGS84 CRS, along with the
    datetime in the UTC time zone. Returns the length of a shadow of an object
    with a specified height at the specified time and location.
    """

mlubej

//VERSION=3
function setup() {
  return {
    input: ["B04", "B08", "CLM"],
    output: { bands: 2 }
  };
}

function evaluatePixel(sample) {
  let ndvi = index(sample.B08, sample.B04);

mlubej

band_names = ['B02', 'B03', 'B04', 'B08', 'B11', 'B12']

add_data = SentinelHubInputTask(
    bands_feature=(FeatureType.DATA, 'BANDS'),
    bands = band_names,
    resolution=10,
    maxcc=0.8,
    time_difference=datetime.timedelta(minutes=120),
    data_source=DataSource.SENTINEL2_L1C,
    additional_data=[(FeatureType.MASK, 'dataMask'),

mlubej

def mate(self, partner):
    """
    Create offspring with a order-crossover method
    :param partner: a second p1 which will participate in the mating
    :return: two new offspring individuals
    """
    c1, c2 = np.zeros((2, self.n_pieces), dtype=int)
    start, end = np.sort(np.random.choice(range(self.n_pieces), 2, replace=False))

    p1 = self.chromosome