def get_noaa_and_image_centroids(image_features_dict, noaa_data):
'''
Output: noaa_cents_labels, tuple
my_cents, nested lists
'''
# For Local Series Computing, uncomment function
#noaa_data = load_noaa_data()
# Cluster/Parallel computing
# noaa_data will have already been pushed into each engine
noaa_cents_labels_sameDay, \
noaa_cents_labels_prevDay=\
Centroid_Labeling.get_currentDay_previusDay_noaa_activeRegions(image_features_dict, noaa_data)
_, hour, minute = image_features_dict["image_time"].split(":")
# check if image occured in 1st or 2nd half of the day
if int(hour) >= 12:
noaa_cents_labels = noaa_cents_labels_sameDay
else:
noaa_cents_labels = noaa_cents_labels_prevDay
my_cents = extract_image_features.get_image_active_region_centroids(image_features_dict, split_centroids = False)
return my_cents, noaa_cents_labels
def plot_sunspots_and_active_regions(df, scan_year, features, time_slice):
'''
use scan_year to shift through noaa observations
use time_slice to scan through image data
'''
noaa_cents, _ = extract.get_noaa_centroids(df, scan_year)
noaa_x, noaa_y = unpack_noaa_cents(noaa_cents)
x_cents, y_cents= extract.get_image_active_region_centroids(features[time_slice])
plt.figure(figsize=(10,10))
noaa = plt.scatter(noaa_x, noaa_y, c='b', marker='o');
me = plt.scatter(x_cents, y_cents , c='r',marker='+');
plt.title("Sunspots & Active Regions " + scan_year);
plt.legend((me, noaa),
('mydata-sunspots', 'noaa-AR'),
scatterpoints=1,
loc='lower right',
ncol=2,
fontsize=15);
