def test(self):
iid = 'LANDSAT/LC08/C01/T1_SR'
collection = ee.ImageCollection(iid)
date = '2017-03-08'
p_cloud = ee.Geometry.Point([-65.7539, -25.0327])
p_clear = ee.Geometry.Point([-66.034, -24.9512])
image = getimage(collection, date, area)
# masked = cloud_mask.cfmask_bits(image)
masked = cloud_mask.landsatSR()(image)
vals = get_value(masked, p_cloud, 30, 'client')
show(iid, image, masked, bands8, 0, 5000)
self.assertEqual(vals["B1"], None)
def test(self):
iid = 'LANDSAT/LE07/C01/T1_SR'
collection = ee.ImageCollection(iid)
date = '2010-03-13'
p_cloud = ee.Geometry.Point([-65.5784, -24.773])
p_clear = ee.Geometry.Point([-65.599, -24.9313])
image = getimage(collection, date, area)
# masked = cloud_mask.cfmask_bits(image)
masked = cloud_mask.landsatSR()(image)
vals = get_value(masked, p_cloud, 30, 'client')
show(iid, image, masked, bands457, 0, 5000)
self.assertEqual(vals["B1"], None)
def test(self):
iid = 'LANDSAT/LT04/C01/T1_SR'
collection = ee.ImageCollection(iid)
date = '1989-07-25'
p_cloud = ee.Geometry.Point([-65.506, -24.9263])
p_clear = ee.Geometry.Point([-64.6133, -24.4522])
image = getimage(collection, date, area)
# masked = cloud_mask.cfmask_bits(image)
masked = cloud_mask.landsatSR()(image)
vals = get_value(masked, p_cloud, 30, 'client')
show(iid, image, masked, bands457, 0, 5000)
self.assertEqual(vals["B1"], None)
def Landsat8USGS(cls):
copy = deepcopy(Collection.Landsat8TOA()) # L8 TOA
copy_usgs = deepcopy(Collection.Landsat5USGS()) # L5 USGS
copy.kws["process"] = copy_usgs.process
copy.kws["max"] = copy_usgs.max
copy.kws["fclouds"] = cld.landsatSR()
# copy.kws["ATM_OP"] = copy_usgs.ATM_OP
copy.kws["equiv"] = "LANDSAT/LC8_L1T_TOA_FMASK"
copy.kws["bandscale"] = copy.bandscale
# copy.kws["col_id"] = 12
copy.kws["short"] = "L8USGS"
copy.kws["clouds_band"] = "pixel_qa"
obj = cls(**copy.kws)
obj.threshold = {
'NIR': {
'min': 700,
'max': 4500
},
'RED': {
'min': 50,
'max': 2000
},
'SWIR': {
'min': 400,
'max': 3500
},
'SWIR2': {
'min': 150,
'max': 2800
},
'BLUE': {
'min': 0,
'max': 1000
},
'GREEN': {
'min': 100,
'max': 1500
}
}
# CAMBIOS
# CHANGE
# obj.ID = "LANDSAT/LC08/C01/T1_SR"
obj.ID = IDS[obj.short]
return obj
def Landsat7USGS(cls):
copy = deepcopy(Collection.Landsat7TOA()) # L5 USGS
copy.kws["equiv"] = "LANDSAT/LE7_L1T_TOA_FMASK"
copy.kws["process"] = "SR"
copy.kws["max"] = 10000
copy.kws["fclouds"] = cld.landsatSR()
copy.kws["ATM_OP"] = "sr_atmos_opacity"
copy.kws["clouds_band"] = "pixel_qa"
# copy.kws["col_id"] = 9
copy.kws["short"] = "L7USGS"
obj = cls(**copy.kws)
obj.threshold = {
'NIR': {
'min': 700,
'max': 4500
},
'RED': {
'min': 50,
'max': 2000
},
'SWIR': {
'min': 400,
'max': 3500
},
'SWIR2': {
'min': 150,
'max': 2800
},
'BLUE': {
'min': 0,
'max': 1000
},
'GREEN': {
'min': 100,
'max': 1500
}
}
# CAMBIO
# CHANGE
# obj.ID = "LANDSAT/LE07/C01/T1_SR"
obj.ID = IDS[obj.short]
return obj
def Landsat4USGS(cls):
copy = deepcopy(Collection.Landsat4TOA())
# copy.kws["col_id"] = 16
copy.kws["short"] = "L4USGS"
copy.kws["max"] = 10000
copy.kws["fclouds"] = cld.landsatSR()
copy.kws["ATM_OP"] = "sr_atmos_opacity"
copy.kws["equiv"] = "LANDSAT/LT4_L1T_TOA_FMASK"
copy.kws["clouds_band"] = "pixel_qa"
obj = cls(**copy.kws)
# CAMBIO
# obj.ID = "LANDSAT/LT04/C01/T1_SR"
obj.ID = IDS[obj.short]
obj.threshold = {
'NIR': {
'min': 700,
'max': 4500
},
'RED': {
'min': 50,
'max': 2000
},
'SWIR': {
'min': 400,
'max': 3500
},
'SWIR2': {
'min': 150,
'max': 2800
},
'BLUE': {
'min': 0,
'max': 1000
},
'GREEN': {
'min': 100,
'max': 1500
}
}
return obj
#reading CSV
dataset = pd.read_csv('dataset.csv')
#dataset=pd.read_csv(r'C:\Users\nsuar\Google Drive\local_testing_files\dataset.csv')
print("dataset read succesfully")
#defining image
startDate = '2020-01-01'
endDate = '2020-12-31'
landsat = ee.ImageCollection("LANDSAT/LC08/C01/T1_SR")
# filter date
landsat = landsat.filterDate(startDate, endDate)
#applying cloud masking
landsat_masked = landsat.map(cloud_mask.landsatSR(['cloud']))
landsat_masked = landsat_masked.select(
["B1", "B2", "B3", "B4", "B5", "B6", "B7"])
# selecting bands for unmasked version
landsat = landsat.select(["B1", "B2", "B3", "B4", "B5", "B6", "B7"])
#list to store the task objects
tasks = []
#running the tasks to get the imagery
N = len(dataset)
batch_size = 100
for j in range(math.ceil(N / batch_size)):
#determining batch lower and upper indexes, given batch size
import ee
import numpy as np
import geetools
from geetools import ui, cloud_mask
import os, datetime
import pandas as pd
import itertools
import tensorflow as tf
cloud_mask_landsatSR = cloud_mask.landsatSR()
cloud_mask_sentinel2 = cloud_mask.sentinel2()
# tfrecord helper functions ----------------------------------------------------
# https://stackoverflow.com/questions/52324515/passing-multiple-inputs-to-keras-model-from-tf-dataset-api
# https://www.tensorflow.org/tutorials/load_data/tfrecord
def _bytes_feature(value):
"""Returns a bytes_list from a string / byte."""
# If the value is an eager tensor BytesList won't unpack a string from an EagerTensor.
if isinstance(value, type(tf.constant(0))):
value = value.numpy()
return tf.train.Feature(bytes_list=tf.train.BytesList(value=[value]))
def _float_feature(value):
"""Returns a float_list from a float / double."""
return tf.train.Feature(float_list=tf.train.FloatList(value=[value]))
def _int64_feature(value):