Skip to content

Time-Varying Ensemble Detection

There are currently two methods of creating time-varying maps.

The first creates maps of varying timescale based on user entered intervals around a specific user inputted center date. These maps are then combined with a weighted average - either user inputted, or evenly weighted.

The code for running average maps is found here.

The second method uses a Gaussian distribution and a user inputted center date. The closer an image is to the center date, the more weight it has. Images are added one at a time to create the resultant maps.

The code for gaussian time weighted maps is found here.

Running Average Maps

EUV Map CHD Map
EUV Map CHD Map

Gaussian Time Varying Maps

EUV Map CHD Map
EUV Map CHD Map

Code Outline

Running Average Maps

This outline creates individual combined maps based on user defined timescales (1 day, 1 week, 2 weeks, etc.), then combines the maps based on user defined weighting. 

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
if timescale_weight is None:
    timescale_weight = [1.0 / len(timescales)] * len(timescales)
weight_sum = sum(timescale_weight)
if weight_sum != 1:
    raise ValueError("The timescale weights do not sum to 1. Please reenter weights or change to None for even weighting of maps.")
lbc_combo_query, iit_combo_query = chd_funcs.get_inst_combos(db_session, inst_list, time_min=max_time_min, time_max=max_time_max)
query_pd = db_funcs.query_euv_images(db_session=db_session, time_min=query_time_min, time_max=query_time_max)
los_image, iit_image, methods_list, use_indices = cr_funcs.apply_ipp(db_session, hdf_data_dir, inst_list, row, methods_list, lbc_combo_query,
                                                                     iit_combo_query, n_intensity_bins=n_intensity_bins, R0=R0)
chd_image = cr_funcs.chd(db_session, inst_list, los_image, iit_image, use_indices, iit_combo_query, thresh1=thresh1, thresh2=thresh2, nc=nc, iters=iters)
euv_map, chd_map = cr_funcs.create_map(iit_image, chd_image, methods_list, row, map_x=map_x, map_y=map_y, R0=R0)
euv_timescale[time_ind], chd_timescale[time_ind], combined_method, chd_combined_method = cr_funcs.cr_map(euv_map, chd_map, euv_timescale[time_ind],
                                 chd_timescale[time_ind], image_info, map_info, mu_cutoff=mu_cutoff, mu_merge_cutoff=mu_merge_cutoff)
euv_combined, chd_combined, timescale_method = dp_funcs.create_timescale_maps(euv_timescale, chd_timescale, timescale_weight, image_info_timescale, map_info_timescale)
dp_funcs.save_timescale_maps(db_session, map_data_dir, euv_combined, chd_combined, image_info_timescale, map_info_timescale, methods_list, combined_method, chd_combined_method, timescale_method)
  • 1.) timescale_weight = [1.0 / len(timescales)] * len(timescales)
    • if the timescale weights are undefined, create an array of even weights
  • 2.) raise ValueError
    • check that the weights in the array add up to 1
  • 3.) chd_funcs.get_inst_combos
    • query the appropriate combo ids for each instrument based off the maximum time range the user inputted
  • 4.) db_funcs.query_euv_images
    • query the database for images based off the time range and center date
  • 5.) cr_funcs.apply_ipp
    • apply image pre-processing corrections
  • 6.) cr_funcs.chd
    • apply the Coronal Hole Detection algorithm to the image
  • 7.) cr_funcs.create_map
    • convert the image and detection to a map
  • 8.) cr_funcs.cr_map
    • create combined maps using the method for synoptic mapping
  • 9.) dp_funcs.create_timescale_maps
  • 10.) dp_funcs.save_timescale_maps
    • plot and save timescale maps to the database, including the methods combination

Gaussian Time Varying Maps

This method first creates a gaussian distribution and then weights images based off their closeness to a center date. 

1
2
3
4
5
6
7
8
lbc_combo_query, iit_combo_query = chd_funcs.get_inst_combos(db_session, inst_list, time_min=query_time_min, time_max=query_time_max)
query_pd = db_funcs.query_euv_images(db_session=db_session, time_min=query_time_min, time_max=query_time_max)
norm_dist = dp_funcs.gauss_time(query_pd, sigma)
los_image, iit_image, methods_list, use_indices = cr_funcs.apply_ipp(db_session, hdf_data_dir, inst_list, row, methods_list, lbc_combo_query, iit_combo_query, n_intensity_bins=n_intensity_bins, R0=R0)
chd_image = cr_funcs.chd(db_session, inst_list, los_image, iit_image, use_indices, iit_combo_query, thresh1=thresh1, thresh2=thresh2, nc=nc, iters=iters)
euv_map, chd_map = cr_funcs.create_map(iit_image, chd_image, methods_list, row, map_x=map_x, map_y=map_y, R0=R0)
euv_combined, chd_combined, sum_wgt, combined_method = dp_funcs.time_wgt_map(euv_map, chd_map, euv_combined, chd_combined, image_info, weight, sum_wgt, sigma, mu_cutoff)
dp_funcs.save_gauss_time_maps(db_session, map_data_dir, euv_combined, chd_combined, image_info, map_info, methods_list, combined_method)
  • 1.) chd_funcs.get_inst_combos
    • query the appropriate combo ids for each instrument based off the maximum time range the user inputted
  • 2.) db_funcs.query_euv_images
    • query the database for images based off the time range and center date
  • 3.) dp_funcs.gauss_time
    • generate a gaussian distribution based off the number of images queried and user defined sigma balue
  • 4.) cr_funcs.apply_ipp
    • apply image pre-processing corrections
  • 5.) cr_funcs.chd
    • apply the Coronal Hole Detection algorithm to the image
  • 6.) cr_funcs.create_map
    • convert the image and detection to a map
  • 6.) dp_funcs.time_wgt_map
    • create combined maps based off the weighting of the Gaussian distribution
  • 7.) dp_funcs.save_gauss_time_maps
    • plot and save time weighted maps to the database, including the methods combination