def convert_nc(yyyymmdd, file_in, out_file):
r = raster.ReadRaster(file_in)
lons = np.linspace(r.xmin + 0.5 * r.xres, r.xmax - 0.5 * r.xres, r.ncol)
lats = np.linspace(r.ymin + np.abs(0.5 * r.yres),
r.ymax - 0.5 * np.abs(r.xres), r.nrow)
rootgrp = Dataset(out_file, "w", format="NETCDF4_CLASSIC")
time = rootgrp.createDimension("time", None)
lat = rootgrp.createDimension("lat", len(lats))
lon = rootgrp.createDimension("lon", len(lons))
times = rootgrp.createVariable("time", "f8", ("time", ))
latitudes = rootgrp.createVariable("lat", "f8", ("lat", ))
longitudes = rootgrp.createVariable("lon", "f8", ("lon", ))
var = rootgrp.createVariable("temp",
"f4", (
"time",
"lat",
"lon",
),
zlib=True,
least_significant_digit=1)
data = np.flipud(r.data)
#data[data>200] = np.nan
latitudes[:] = lats
longitudes[:] = lons
var[0, :, ] = data
rootgrp.description = "Computed by Satyukt Analytics"
rootgrp.history = "Created %s " % dt.date.today()
rootgrp.source = "Estimated using vikleda"
latitudes.units = "degrees north"
longitudes.units = "degrees east"
times.units = "days since 0001-01-01"
times.calendar = "gregorian"
dates = dt.datetime.strptime(yyyymmdd, '%Y%m%d')
times[0] = date2num(dates, units=times.units, calendar=times.calendar)
rootgrp.close()
print('%s file created' % out_file)
def convert_nc(yyyymmdd, dir_tif, dir_nc):
files_in = glob.glob(os.path.join(dir_tif, "S1?_??_????_????_%s*.tif"%yyyymmdd))
for file_in in files_in:
out_file = os.path.join(dir_nc, os.path.basename(file_in).replace(".tif",".nc"))
if not os.path.exists(out_file):
r = raster.ReadRaster(file_in)
lons = np.linspace(r.xmin+0.5*r.xres, r.xmax-0.5*r.xres, r.ncol)
lats = np.linspace(r.ymin+np.abs(0.5*r.yres), r.ymax-0.5*np.abs(r.xres), r.nrow)
rootgrp = Dataset(out_file, "w", format="NETCDF4_CLASSIC")
time = rootgrp.createDimension("time", None)
lat = rootgrp.createDimension("lat", len(lats))
lon = rootgrp.createDimension("lon", len(lons))
times = rootgrp.createVariable("time","f8",("time",))
latitudes = rootgrp.createVariable("lat","f8",("lat",))
longitudes = rootgrp.createVariable("lon","f8",("lon",))
var = rootgrp.createVariable("temp","f4",("time","lat","lon",), zlib=True,
least_significant_digit=1)
data = np.flipud(r.data)
if "VV" in file_in or "VH" in file_in:
data = 10*np.log10(data)
#data[data>200] = np.nan
latitudes[:] = lats
longitudes[:] = lons
var[0,:,] = data
rootgrp.description = "Computed by Satyukt Analytics"
rootgrp.history = "Created %s "%dt.date.today()
rootgrp.source = "Estimated using vikleda"
latitudes.units = "degrees north"
longitudes.units = "degrees east"
times.units = "days since 0001-01-01"
times.calendar = "gregorian"
dates = dt.datetime.strptime(yyyymmdd,'%Y%m%d')
times[0] = date2num(dates, units=times.units, calendar=times.calendar)
rootgrp.close()
def store_evolution_in(lst):
"""Returns a callback function to store the evolution of the level sets in
the given list.
"""
def _store(x):
lst.append(np.copy(x))
return _store
in_dir = "/home/pavithra/Desktop/Satyukt/20th Dec 2018_B2.tif"
sd_dir = "/home/pavithra/Desktop/New_test_folder"
r_band = raster.ReadRaster(in_dir)
#data = r_band.data
image = r_band.data
init_ls = checkerboard_level_set(image.shape, 5)
evolution = []
callback = store_evolution_in(evolution)
ls = morphological_chan_vese(image,
100,
init_level_set=init_ls,
smoothing=5,
iter_callback=callback)
fig, axes = plt.subplots(1, 2, figsize=(8, 8))
import scipy
from apya import raster
from scipy import misc
from scipy.cluster.vq import kmeans, vq, whiten
from skimage import filters
in_dir = "/home/pavithra/Desktop/Rasterimages/AndhraPradesh_SCL/*SCL_20m.tif"
in_files = glob.glob(in_dir)
in_files.sort()
sd_dir = "/home/pavithra/Desktop/Rasterimages/AndhraPradesh_Data/Cloudy/"
sd_dir2 = "/home/pavithra/Desktop/Rasterimages/AndhraPradesh_Data/clear_data/"
for in_file in in_files:
reading = raster.ReadRaster(in_file)
im = Image.open(in_file, 'r')
pix_val = list(im.getdata())
#getting the count of values
five = pix_val.count(5.0)
seven = pix_val.count(7.0)
four = pix_val.count(4.0)
three = pix_val.count(3.0)
#print('5',five)
#print('7',seven)
#print('4',four)
#print('3',three)
#total pixel count
#This program coverts the featured images into a different pixel composition. Replacing the black and grey ones with white and the white with black, to make the boundaries more clear, to implement the algorithm.
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from scipy import ndimage as ndi
import cv2
from skimage.feature import shape_index
from apya import raster
from PIL import Image
import os
import cv2
im = raster.ReadRaster(
'/home/pavithra/Desktop/Rasterimages/NEW_Feature_data/B02/T14QQF_20181220T165719_B02_10m.tif'
)
sd_dir = "/home/pavithra/Desktop/Satyukt"
data = im.data
print(data[1][2])
print(data.shape)
#955 is the number of pixels in row and 1190 in the column
for i in range(0, 955):
for j in range(0, 1190):
if (data[i][j] <= 65):
data[i][j] = 255
from apya import raster
from scipy import misc
from scipy.cluster.vq import kmeans, vq, whiten
from skimage import filters
from skimage import measure
#finding the Standard deviation of the bands
#For cloudy data, the 'less cloudy' data is applied feature extraction
in_dir = "/home/pavithra/Desktop/Rasterimages/NEW_MEX_DATA/mex_data/B12/*.tif"
in_files = glob.glob(in_dir)
in_files.sort()
sd_dir = "/home/pavithra/Desktop/Rasterimages/NEW_SD/SD_B12"
for in_file in in_files:
r_band = raster.ReadRaster(in_file)
data = r_band.data
out_sd = ndimage.generic_filter(data, np.var, size=3)**0.5
sd_file = os.path.join(sd_dir, os.path.basename(in_file))
r_band_sd = raster.Raster(out_sd,
extent=r_band.extent,
projection=r_band.projection)
raster.WriteRaster(r_band_sd, sd_file)
print(sd_file)
#Applying Guassian filter
sd_dir = "/home/pavithra/Desktop/Rasterimages/AndhraPradesh_Data/*.tif"
sd_files = glob.glob(sd_dir)
sd_files.sort()
sd_datacc = "/home/pavithra/Desktop/Rasterimages/AndhraPradesh_Data/Data_cc/"
for in_file in in_files:
for ik_file in sd_files:
one = os.path.basename(in_file)
on = one.rsplit('_',2)
two = os.path.basename(ik_file)
tw = two.rsplit('_',2)
if(on[0] == tw[0]):
r_band = raster.ReadRaster(ik_file)
data = r_band.data
#data1 = r_band.astype(np.uint8)
for i in range(len(data)):
for j in range(len(data[0])):
if(data[i,j]>3000):
data[i,j] = np.nan
else:
data[i,j] = data[i,j]
sd_file = os.path.join(sd_datacc, os.path.basename(ik_file))
r_band_cc = raster.Raster(data, extent=r_band.extent, projection=r_band.projection)
raster.WriteRaster(r_band_cc, sd_file)
print(sd_file)
else:
print('file not matched')
import cv2 import argparse import os from apya import raster import matplotlib.pyplot as plt #destination folder sd_dir = '/home/pavithra/Desktop/New_test_folder/Canny_level2_outputs/' #input image image = '/home/pavithra/Desktop/Rasterimages/Test_pixel/T44QND_20190317T045659_B03_10m.tif' #Reading the raster data read_rast = raster.ReadRaster(image) data_canni = read_rast.data img = cv2.imread(image,cv2.IMREAD_UNCHANGED) img = (img).astype(np.uint8) print(type(img)) canny = cv2.Canny(img,100,200) print(canny) titles = ['Band-8', 'Output'] images = [img, (canny)]