def test_rasterize_vector_and_write_to_disk(self):
from madmex.mapper.data.raster import new_options_for_create_raster_from_reference, create_raster_tiff_from_reference
from madmex.core.controller.commands import get_bundle_from_path
from madmex.mapper.data import raster
BUNDLE_PACKAGE = 'madmex.mapper.bundle'
landmask_path = '/Users/erickpalacios/Documents/CONABIO/MADMEXdata/eodata/footprints/country_mexico/'
bundle = get_bundle_from_path(landmask_path, '../../../mapper',
BUNDLE_PACKAGE)
extents_dictionary = {
u'x_range': 7521.0,
u'y_range': 7741.0,
u'properties': {
u'projection':
'PROJCS["UTM Zone 15, Northern Hemisphere",GEOGCS["Unknown datum based upon the WGS 84 ellipsoid",DATUM["Not specified (based on WGS 84 spheroid)",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]]],PRIMEM["Greenwich",0],UNIT["degree",0.0174532925199433]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",-93],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",0],UNIT["Meter",1]]',
u'geotransform': (523185.0, 30.0, 0.0, 2033715.0, 0.0, -30.0)
},
u'x_offset': 'array([ 10., 30., 0., 50.])',
u'y_offset': 'array([-0., -0., -0., -0.])'
}
options_to_create = new_options_for_create_raster_from_reference(
extents_dictionary, raster.DATA_SHAPE,
(int(extents_dictionary['x_range']),
int(extents_dictionary['y_range']), 1), {})
image = '/Users/erickpalacios/Documents/CONABIO/Tareas/Redisenio_MADMEX/clasificacion_landsat/landsat8/classification/rasterize2.tif'
#The next line create a tiff with empty array
data = create_raster_tiff_from_reference(extents_dictionary, image,
None, options_to_create)
bundle.rasterize(data, [1], [1])
new_options_for_create_raster_from_reference(extents_dictionary,
raster.DATASET, data,
options_to_create)
create_raster_tiff_from_reference(extents_dictionary, image,
data.ReadAsArray()) #created in disk
def test_create_raster_in_memory_and_then_write_to_disk(self):
from madmex.mapper.data.raster import new_options_for_create_raster_from_reference, create_raster_tiff_from_reference
from madmex.mapper.data import raster
extents_dictionary = {
u'x_range': 7521.0,
u'y_range': 7741.0,
u'properties': {
u'projection':
'PROJCS["UTM Zone 15, Northern Hemisphere",GEOGCS["Unknown datum based upon the WGS 84 ellipsoid",DATUM["Not specified (based on WGS 84 spheroid)",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]]],PRIMEM["Greenwich",0],UNIT["degree",0.0174532925199433]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",-93],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",0],UNIT["Meter",1]]',
u'geotransform': (523185.0, 30.0, 0.0, 2033715.0, 0.0, -30.0)
},
u'x_offset': 'array([ 10., 30., 0., 50.])',
u'y_offset': 'array([-0., -0., -0., -0.])'
}
options_to_create = new_options_for_create_raster_from_reference(
extents_dictionary, raster.DATA_SHAPE,
(int(extents_dictionary['x_range']),
int(extents_dictionary['y_range']), 1), {})
data = create_raster_tiff_from_reference(
extents_dictionary, '', None,
options_to_create) #created in memory
new_options_for_create_raster_from_reference(extents_dictionary,
raster.DATASET, data,
options_to_create)
image = '/Users/erickpalacios/Documents/CONABIO/Tareas/Redisenio_MADMEX/clasificacion_landsat/landsat8/classification/raster_test_created.tif'
create_raster_tiff_from_reference(extents_dictionary, image,
data.ReadAsArray()) #created in disk
def test_create_raster_in_memory_and_then_write_to_disk(self):
from madmex.mapper.data.raster import new_options_for_create_raster_from_reference, create_raster_tiff_from_reference
from madmex.mapper.data import raster
extents_dictionary = {u'x_range': 7521.0, u'y_range': 7741.0, u'properties': {u'projection': 'PROJCS["UTM Zone 15, Northern Hemisphere",GEOGCS["Unknown datum based upon the WGS 84 ellipsoid",DATUM["Not specified (based on WGS 84 spheroid)",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]]],PRIMEM["Greenwich",0],UNIT["degree",0.0174532925199433]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",-93],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",0],UNIT["Meter",1]]', u'geotransform': (523185.0, 30.0, 0.0, 2033715.0, 0.0, -30.0)}, u'x_offset': 'array([ 10., 30., 0., 50.])', u'y_offset': 'array([-0., -0., -0., -0.])'}
options_to_create = new_options_for_create_raster_from_reference(extents_dictionary, raster.DATA_SHAPE, (int(extents_dictionary['x_range']), int(extents_dictionary['y_range']), 1), {})
data = create_raster_tiff_from_reference(extents_dictionary, '', None, options_to_create) #created in memory
new_options_for_create_raster_from_reference(extents_dictionary, raster.DATASET, data, options_to_create)
image = '/Users/erickpalacios/Documents/CONABIO/Tareas/Redisenio_MADMEX/clasificacion_landsat/landsat8/classification/raster_test_created.tif'
create_raster_tiff_from_reference(extents_dictionary, image, data.ReadAsArray()) #created in disk
def test_rasterize_vector_and_write_to_disk(self):
from madmex.mapper.data.raster import new_options_for_create_raster_from_reference, create_raster_tiff_from_reference
from madmex.core.controller.commands import get_bundle_from_path
from madmex.mapper.data import raster
BUNDLE_PACKAGE = 'madmex.mapper.bundle'
landmask_path = '/Users/erickpalacios/Documents/CONABIO/MADMEXdata/eodata/footprints/country_mexico/'
bundle = get_bundle_from_path(landmask_path, '../../../mapper', BUNDLE_PACKAGE)
extents_dictionary = {u'x_range': 7521.0, u'y_range': 7741.0, u'properties': {u'projection': 'PROJCS["UTM Zone 15, Northern Hemisphere",GEOGCS["Unknown datum based upon the WGS 84 ellipsoid",DATUM["Not specified (based on WGS 84 spheroid)",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]]],PRIMEM["Greenwich",0],UNIT["degree",0.0174532925199433]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",-93],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",0],UNIT["Meter",1]]', u'geotransform': (523185.0, 30.0, 0.0, 2033715.0, 0.0, -30.0)}, u'x_offset': 'array([ 10., 30., 0., 50.])', u'y_offset': 'array([-0., -0., -0., -0.])'}
options_to_create = new_options_for_create_raster_from_reference(extents_dictionary, raster.DATA_SHAPE, (int(extents_dictionary['x_range']), int(extents_dictionary['y_range']), 1), {})
image = '/Users/erickpalacios/Documents/CONABIO/Tareas/Redisenio_MADMEX/clasificacion_landsat/landsat8/classification/rasterize2.tif'
#The next line create a tiff with empty array
data = create_raster_tiff_from_reference(extents_dictionary, image, None, options_to_create)
bundle.rasterize(data, [1], [1])
new_options_for_create_raster_from_reference(extents_dictionary, raster.DATASET, data, options_to_create)
create_raster_tiff_from_reference(extents_dictionary, image, data.ReadAsArray()) #created in disk
def handle(self, **options):
image_to_be_classified = options['image'][0]
#landmask_path = options['landmask_path'][0]
outlier = options['outlier'][0]
folder_results = getattr(SETTINGS, 'BIG_FOLDER')
shutil.copy(image_to_be_classified, folder_results)
image_to_be_classified = folder_results + get_basename_of_file(image_to_be_classified)
landmask_path = getattr(SETTINGS, 'LANDMASK_PATH')
image_for_segmentation = '/results/' + get_basename_of_file(image_to_be_classified)
LOGGER.info('Starting segmentation with: %s' % image_for_segmentation)
val_t = 50
val_s = 0.7
val_c = 0.3
val_xt = 40
val_rows = 625
val_tile = True
val_mp = True
folder_and_bind_segmentation = getattr(SETTINGS, 'FOLDER_SEGMENTATION')
folder_and_bind_license = getattr(SETTINGS, 'FOLDER_SEGMENTATION_LICENSE')
folder_and_bind_image = getattr(SETTINGS, 'BIG_FOLDER_HOST')
LOGGER.info('starting segmentation')
command = 'run_container'
hosts_from_command = get_host_from_command(command)
LOGGER.info('The command to be executed is %s in the host %s' % (command, hosts_from_command[0].hostname))
remote = RemoteProcessLauncher(hosts_from_command[0])
arguments = 'docker run --rm -v ' + folder_and_bind_segmentation + ' -v ' + folder_and_bind_license + ' -v ' + folder_and_bind_image + ' madmex/segmentation python /segmentation/segment.py ' + image_for_segmentation
arguments+= ' -t ' + str(val_t) + ' -s ' + str(val_s) + ' -c ' + str(val_c) + ' --tile ' + str(val_tile) + ' --mp ' + str(val_mp) + ' --xt ' + str(val_xt) + ' --rows ' + str(val_rows)
remote.execute(arguments)
LOGGER.info('Finished segmentation')
image_segmentation_file = image_to_be_classified + '_' + str(val_t) + '_' + ''.join(str(val_s).split('.'))+ '_' + ''.join(str(val_c).split('.')) + '.tif'
LOGGER.info('Starting vectorization of segmentation file: %s' % image_segmentation_file)
image_segmentation_shp_file = image_segmentation_file + '.shp'
vectorize_raster(image_segmentation_file, 1, image_segmentation_shp_file, 'objects', 'id')
LOGGER.info('Finished vectorization: %s' % image_segmentation_shp_file)
gdal_format = 'GTiff'
image_to_be_classified_class = raster.Data(image_to_be_classified, gdal_format)
width, height, bands = image_to_be_classified_class.get_attribute(raster.DATA_SHAPE)
LOGGER.info('Identifying landmask %s' % landmask_path)
bundle = _get_bundle_from_path(landmask_path)
if bundle:
LOGGER.info('Directory %s is a %s bundle', landmask_path, bundle.get_name())
LOGGER.info('Rasterizing vector shape')
options_to_create = new_options_for_create_raster_from_reference(image_to_be_classified_class.metadata, raster.DATA_SHAPE, (width,height, 1), {})
dataset_landmask_rasterized = create_raster_tiff_from_reference(image_to_be_classified_class.metadata, '', None, options_to_create)
bundle.rasterize(dataset_landmask_rasterized, [1], [1]) #the rasterized process changes the dataset
options_to_create = new_options_for_create_raster_from_reference(image_to_be_classified_class.metadata, raster.GDAL_CREATE_OPTIONS, ['COMPRESS=LZW'], {})
image = folder_results + 'landmask_rasterized.tif'
create_raster_tiff_from_reference(image_to_be_classified_class.metadata, image, dataset_landmask_rasterized.ReadAsArray(), options_to_create)
LOGGER.info('Finished rasterizing vector shape')
LOGGER.info('Polygonizing the landmask rasterized')
landmask_folder = folder_results + 'landmask_from_rasterize/'
layer_landmask = 'landmask'
landmask_file = landmask_folder + layer_landmask + '.shp'
layer_landmask = 'landmask'
vectorize_raster(folder_results + 'landmask_rasterized.tif', 1, landmask_folder, layer_landmask, 'id')
LOGGER.info('Folder of polygon: %s' % landmask_folder)
image_segmentation_file_class = raster.Data(image_segmentation_file, gdal_format)
LOGGER.info('Reading array of %s' % image_for_segmentation)
array_sg_raster = image_segmentation_file_class.read_data_file_as_array()
unique_labels_for_objects = numpy.unique(array_sg_raster)
LOGGER.info('Calculating zonal stats for :%s' % image_to_be_classified)
LOGGER.info('Reading array of %s' % image_to_be_classified)
array_image_to_be_classified = image_to_be_classified_class.read_data_file_as_array()
array_zonal_statistics = calculate_zonal_statistics(array_image_to_be_classified, array_sg_raster, unique_labels_for_objects)
LOGGER.info('finished zonal statistics')
array_zonal_statistics_labeled = append_labels_to_array(array_zonal_statistics, unique_labels_for_objects)
LOGGER.info('Shape of array of zonal statistics labeled %s %s' % (array_zonal_statistics_labeled.shape[0], array_zonal_statistics_labeled.shape[1]))
LOGGER.info('Building data frame')
dataframe_zonal_statistics = create_names_of_dataframe_from_filename(build_dataframe_from_array(array_zonal_statistics_labeled.T), array_zonal_statistics_labeled.shape[0], get_basename_of_file(image_to_be_classified))
LOGGER.info('Filling NaN with zeros')
dataframe_zonal_statistics = dataframe_zonal_statistics.fillna(0)
file_name = folder_results + 'dataframe_zonal_statistics'
dataframe_zonal_statistics.to_csv(file_name, sep='\t', encoding='utf-8', index = False)
LOGGER.info('Working with the training data')
training_data_file = getattr(SETTINGS, 'TRAINING_DATA')
LOGGER.info('Clipping training_data_file: %s with: %s' % (training_data_file, landmask_file))
training_data_file_clipped = folder_results + get_basename_of_file(training_data_file) + '_cropped_subprocess_call.tif'
command = [
'gdalwarp', '-cutline', landmask_file,
'-crop_to_cutline', '-of', 'GTiff','-co', 'compress=lzw', '-co', 'tiled=yes', training_data_file, training_data_file_clipped
]
subprocess.call(command)
LOGGER.info('Finished clipping of training data file')
LOGGER.info('Starting warping of file: %s according to %s ' % (training_data_file_clipped, image_segmentation_file))
dataset_warped_training_data_file = warp_raster_from_reference(training_data_file_clipped, image_segmentation_file_class.data_file, None)
LOGGER.info('Starting resizing of array of training file: %s' % training_data_file_clipped)
array_resized_and_warped_training_data_file = get_array_resized_from_reference_dataset(dataset_warped_training_data_file, image_segmentation_file_class.data_file)
import gdal
training_data_file_resized_and_warped = folder_results + get_basename_of_file(training_data_file) + '_resized_and_warped.tif'
options_to_create = new_options_for_create_raster_from_reference(image_to_be_classified_class.metadata, raster.GDAL_CREATE_OPTIONS, ['TILED=YES', 'COMPRESS=LZW', 'INTERLEAVE=BAND'], {})
create_raster_tiff_from_reference(image_to_be_classified_class.metadata, training_data_file_resized_and_warped, array_resized_and_warped_training_data_file, options_to_create, data_type = gdal.GDT_Int32)
LOGGER.info('Starting resampling')
array_training_data_resampled = resample_numpy_array(array_resized_and_warped_training_data_file, width, height, interpolation = 'nearest')
training_data_file_resampled = folder_results + get_basename_of_file(training_data_file) + '_resampled_from_resized_and_warped.tif'
create_raster_tiff_from_reference(image_to_be_classified_class.metadata, training_data_file_resampled, array_training_data_resampled, options_to_create, data_type = gdal.GDT_Int32)
LOGGER.info('Calculating zonal histograms for file: %s according to: %s' % (training_data_file_resampled, image_segmentation_file))
unique_classes = numpy.unique(array_training_data_resampled)
array_of_distribution_of_classes_per_object_segmentation = calculate_zonal_histograms(array_training_data_resampled, unique_classes, array_sg_raster, unique_labels_for_objects)
LOGGER.info('Shape of zonal histogram: %s %s' % (array_of_distribution_of_classes_per_object_segmentation.shape[0], array_of_distribution_of_classes_per_object_segmentation.shape[1]))
array_training_data_resampled = None
LOGGER.info('Getting objects that have a class of at least .75 proportion within zonal histogram')
dataframe_of_objects_for_training_data = get_objects_by_relative_proportion_from_raster_as_dataframe(array_of_distribution_of_classes_per_object_segmentation, unique_labels_for_objects, unique_classes, ["id", "given"], 0.75)
file_name = folder_results + 'dataframe_of_objects_for_training_data'
dataframe_of_objects_for_training_data.to_csv(file_name, sep='\t', encoding='utf-8', index = False)
LOGGER.info('Number of rows and columns of dataframe of pure objects of training data %s %s' % (len(dataframe_of_objects_for_training_data.index), len(dataframe_of_objects_for_training_data.columns) ))
array_of_distribution_of_classes_per_object_segmentation = None
dataframe_of_objects_for_training_data = None
dataframe_of_objects_for_training_data = pandas.read_csv(file_name, sep='\t')
LOGGER.info('Joining dataframe of dataframe zonal statistics and dataframe of objects of training data')
dataframe_all_joined_classified = join_dataframes_by_column_name([dataframe_zonal_statistics, dataframe_of_objects_for_training_data], 'id')
LOGGER.info('Number of rows and columns of dataframe joined %s %s' % (len(dataframe_all_joined_classified.index), len(dataframe_all_joined_classified.columns) ))
if outlier == 'True':
LOGGER.info('Starting outlier elimination with dataframe of zonal statistics and dataframe of pure objects of training data')
LOGGER.info('Starting principal component analysis')
array_reduced_pca = reduce_dimensionality(dataframe_all_joined_classified, .95, ['id', 'given'])
LOGGER.info('Shape of reduced array of zonal statistics and pure objects of training data by pca: %s %s' %(array_reduced_pca.shape[0], array_reduced_pca.shape[1]) )
labels_of_objects_reduced_dataframe = dataframe_all_joined_classified['id'].values
LOGGER.info('Appending labels')
array_reduced_pca_labeled = append_labels_to_array(array_reduced_pca.T, labels_of_objects_reduced_dataframe)
LOGGER.info('Shape of array reduced by pca and labeled: %s %s' %(array_reduced_pca_labeled.shape[0], array_reduced_pca_labeled.shape[1]))
LOGGER.info('Building data frame')
dataframe_reduced_pca_file = folder_results + 'dataframe_joined_for_zonal_statistics_and_pure_objects_of_training_data_reduced_by_pca'
dataframe_reduced_pca = create_names_of_dataframe_from_filename(build_dataframe_from_array(array_reduced_pca_labeled.T), array_reduced_pca_labeled.shape[0], get_basename_of_file(dataframe_reduced_pca_file))
dataframe_reduced_pca.to_csv(dataframe_reduced_pca_file, sep=',', encoding='utf-8', index = False)
LOGGER.info('Starting with elimination of outliers')
LOGGER.info('Joining reduced dataframe by pca with object ids and dataframe of pure objects of training data')
dataframe_reduced_pca_with_classes= join_dataframes_by_column_name([dataframe_reduced_pca, dataframe_of_objects_for_training_data], 'id')
LOGGER.info('Number of rows and columns of dataframe joined: (%s,%s)' %(len(dataframe_reduced_pca_with_classes.index), len(dataframe_reduced_pca_with_classes.columns)))
dataframe_reduced_pca_with_classes.to_csv(dataframe_reduced_pca_file + 'classes', sep = ',', encoding = 'utf8', index = False)
unique_classes = numpy.unique(dataframe_of_objects_for_training_data['given'].values)
object_ids_outlier_elimination = outlier_elimination_for_dataframe(dataframe_reduced_pca_with_classes, 'id', 'given', 'id', 3, unique_classes, 0.15)
object_ids_outlier_elimination_file = folder_results + 'dataframe_object_ids_outlier_elimination'
object_ids_outlier_elimination.to_csv(object_ids_outlier_elimination_file, sep = ',', encoding = 'utf-8', index = False)
LOGGER.info('Joining all dataframes according to ids of outlier elimination ')
dataframe_all_joined_classified = join_dataframes_by_column_name([object_ids_outlier_elimination, dataframe_all_joined_classified], 'id')
LOGGER.info('Number of rows and columns of dataframe joined classified: (%s,%s)' %(len(dataframe_all_joined_classified.index), len(dataframe_all_joined_classified.columns)))
dataframe_zonal_statistics['given'] = '?'
LOGGER.info('Number of rows and columns of dataframe for classifying: (%s,%s)' %(len(dataframe_zonal_statistics.index), len(dataframe_zonal_statistics.columns)))
index_of_objects_not_id_zero = dataframe_zonal_statistics['id'] > 0
dataframe_all_joined_for_classifying = dataframe_zonal_statistics[index_of_objects_not_id_zero]
LOGGER.info('Number of rows and columns of dataframe for classifying after removing object with id zero: (%s,%s)' %(len(dataframe_all_joined_for_classifying.index), len(dataframe_all_joined_for_classifying.columns)))
LOGGER.info('Generating data file')
dataframe_all_joined_classified_file = folder_results + 'C5.data'
dataframe_all_joined_classified.to_csv(dataframe_all_joined_classified_file, sep = ',', encoding = 'utf-8', index = False, header = False)
LOGGER.info('Generating cases file')
dataframe_all_joined_for_classifying_file = folder_results + 'C5.cases'
dataframe_all_joined_for_classifying.to_csv(dataframe_all_joined_for_classifying_file, sep = ',', encoding = 'utf-8', index = False, header = False)
LOGGER.info('Generating names file')
unique_classes = numpy.unique(dataframe_all_joined_classified['given'].values)
name_namesfile = folder_results + 'C5.names'
generate_namesfile(dataframe_all_joined_classified.columns, unique_classes,name_namesfile, 'id', 'given')
command = 'run_container'
hosts_from_command = get_host_from_command(command)
LOGGER.info('The command to be executed is %s in the host %s' % (command, hosts_from_command[0].hostname))
remote = RemoteProcessLauncher(hosts_from_command[0])
folder_and_bind_c5 = getattr(SETTINGS, 'BIG_FOLDER_HOST')
arguments = 'docker run --rm -v ' + folder_and_bind_c5 + ' madmex/c5_execution ' + 'c5.0 -b -f /results/C5'
LOGGER.info('Beginning C5')
remote.execute(arguments)
LOGGER.info('Begining predict')
arguments = 'docker run --rm -v ' + folder_and_bind_c5 + ' madmex/c5_execution ' + 'predict -f /results/C5'
remote = RemoteProcessLauncher(hosts_from_command[0])
output = remote.execute(arguments, True)
LOGGER.info('Writing C5 result to csv')
C5_result = write_C5_result_to_csv(output, folder_results)
LOGGER.info('Using result of C5: %s for generating land cover shapefile and raster image' % C5_result)
LOGGER.info('Using result of C5 for generating land cover shapefile and raster image')
C5_result = folder_results + 'C5_result.csv'
dataframe_c5_result = pandas.read_csv(C5_result)
FORMAT = 'ESRI Shapefile'
image_segmentation_shp_class = vector.Data(image_segmentation_shp_file, FORMAT)
LOGGER.info('Joining dataframe %s to %s' %(C5_result, image_segmentation_shp_file))
dataframe_joined_shp_segmentation_and_c5_result = join_C5_dataframe_and_shape(image_segmentation_shp_class, 'id', dataframe_c5_result, 'id')
LOGGER.info('Number of rows and columns of dataframe joined: (%s,%s)' %(len(dataframe_joined_shp_segmentation_and_c5_result.index), len(dataframe_joined_shp_segmentation_and_c5_result.columns)))
dataframe_joined_shp_segmentation_and_c5_result_file = folder_results + 'dataframe_joined_shp_segmentation_and_c5_result.csv'
LOGGER.info('Writing csv of join between c5 result and segmentation shape: %s' % dataframe_joined_shp_segmentation_and_c5_result_file)
dataframe_joined_shp_segmentation_and_c5_result.to_csv(dataframe_joined_shp_segmentation_and_c5_result_file, sep =',', encoding = 'utf8', index = False)
LOGGER.info('Writing C5 result joined with segmentation shape to shapefile')
segmentation_and_c5_result_file_vectorized_folder = folder_results + 'segmentation_and_c5_result_vectorized/'
create_directory_path(segmentation_and_c5_result_file_vectorized_folder)
sql = "SELECT a.id, a.predicted, a.confidence, st_geomfromtext(a.geom," + image_segmentation_shp_class.srid+ ") as geometry "
sql+= "from dataframe_joined_shp_segmentation_and_c5_result a"
shp_result = segmentation_and_c5_result_file_vectorized_folder + '/C5_result_joined_segmentation_shape.shp'
command = [
'ogr2ogr', shp_result,
dataframe_joined_shp_segmentation_and_c5_result_file,
'-dialect', 'sqlite', '-sql', sql
]
subprocess.call(command)
LOGGER.info('Rasterizing segmentation and c5 result shape of folder %s' % segmentation_and_c5_result_file_vectorized_folder)
LOGGER.info('Identifying segmentation and c5 shape folder %s' % segmentation_and_c5_result_file_vectorized_folder)
bundle = _get_bundle_from_path(segmentation_and_c5_result_file_vectorized_folder)
if bundle:
LOGGER.info('Directory %s is a %s bundle', segmentation_and_c5_result_file_vectorized_folder, bundle.get_name())
LOGGER.info('Rasterizing vector shape to get land cover tif')
options_to_create = new_options_for_create_raster_from_reference(image_to_be_classified_class.metadata, raster.DATA_SHAPE, (width,height, 1), {})
dataset_shape_sg_and_c5_rasterized = create_raster_tiff_from_reference(image_to_be_classified_class.metadata, '', None, options_to_create)
bundle.rasterize(dataset_shape_sg_and_c5_rasterized, [1], None, ["ATTRIBUTE=predicted" ]) #the rasterized process changes the dataset
options_to_create = new_options_for_create_raster_from_reference(image_to_be_classified_class.metadata, raster.GDAL_CREATE_OPTIONS, ['COMPRESS=LZW'], {})
image =folder_results + 'madmex_lcc_prueba.tif'
create_raster_tiff_from_reference(image_to_be_classified_class.metadata, image, dataset_shape_sg_and_c5_rasterized.ReadAsArray(), options_to_create, data_type = gdal.GDT_Int32)
LOGGER.info('Finished rasterizing vector shape')
LOGGER.info('Rasterizing vector shape to get confidence tif')
bundle.rasterize(dataset_shape_sg_and_c5_rasterized, [1], None, ["ATTRIBUTE=confidence" ])
image =folder_results + 'madmex_lcc_confidence_prueba.tif'
create_raster_tiff_from_reference(image_to_be_classified_class.metadata, image, dataset_shape_sg_and_c5_rasterized.ReadAsArray(), options_to_create)
LOGGER.info('Finished rasterizing vector shape')
LOGGER.info('Finished workflow classification :)')
