Updating detection

The dieback detection can be updated simply by adding new Sentinel-2 dates which can be very useful for monitoring vegetation. To illustrate this, copy the files from the directory update_study_area to the directory study_area, then run the entire script again. You can save results from the Results, Timelapses and TimeSeries in an other folder if you wish to compare them with the updated results which will overwrite them.

Here, you can find the complete script all in one place :

from fordead.steps.step1_compute_masked_vegetationindex import compute_masked_vegetationindex
from fordead.steps.step2_train_model import train_model
from fordead.steps.step3_dieback_detection import dieback_detection
from fordead.steps.step4_compute_forest_mask import compute_forest_mask
from fordead.steps.step5_export_results import export_results

from fordead.visualisation.create_timelapse import create_timelapse
from fordead.visualisation.vi_series_visualisation import vi_series_visualisation

input_directory = "<MyWorkingDirectory>/study_area"
data_directory = "<output directory>"

compute_masked_vegetationindex(input_directory = input_directory, 
                               data_directory = data_directory, 
                               lim_perc_cloud = 0.4, 
                               interpolation_order = 0, 
                               sentinel_source  = "THEIA", 
                               soil_detection = False, 
                               formula_mask = "B2 > 600", 
                               vi = "CRSWIR", 
                               apply_source_mask = True)

train_model(data_directory = data_directory, 
            nb_min_date = 10, 
            min_last_date_training="2018-01-01", 
            max_last_date_training="2018-06-01")

dieback_detection(data_directory = data_directory, 
                  threshold_anomaly = 0.16,
                  stress_index_mode = "weighted_mean")

compute_forest_mask(data_directory, 
                    forest_mask_source = "vector", 
                    vector_path = "<MyWorkingDirectory>/vector/area_interest.shp")

export_results(data_directory = data_directory, 
               frequency= "M", 
               multiple_files = False, 
               conf_threshold_list = [0.265],
               conf_classes_list = ["Low anomaly","Severe anomaly"])

create_timelapse(data_directory = data_directory, 
                 x = 643069, 
                 y = 5452565, 
                 buffer = 1500)

vi_series_visualisation(data_directory = data_directory, 
                        shape_path = "<MyWorkingDirectory>/vector/points_for_graphs.shp", 
                        name_column = "id", 
                        ymin = 0, 
                        ymax = 2, 
                        chunks = 100)

Results can only be updated using Sentinel-2 acquisitons more recent than those previously processed.

The following figures show how the bark beetle attack progressed for the entire area, and for a single pixel :

Original	Updated