Use Wavelet Denoising to find transients in CTA DL3 data. This data consists of 2d histograms containing number of gammalike events per ra/dec bin. Each histogram represents a short duration of observation (around 30 seconds). In this case a steady source (crab), a number of transients and the cosmic ray background is simulated. Wavelet Denoising is used to clean the images/histograms from background-noise (see animation below). From the cleaned imgages a trigger criterion is extracted and monitored over time to find objects with high variability in CTA data.
make Makefile settings:
- irf path - path to CTA's IRF simulations
- n_transients = Number simulated transients
- slices_per_part = Number timesteps for each simulation step: 1 transient = 1 simulation without transient, 1 simulation with transient, one simulation without transinet = 3*slices_per_part for 1 transient
- transient_template_filename = random if exponential or gaussian, else 0 or 1 for gaussian an 2 for exponential
- threshold = threshold for the trigger criterion in a.u.
- Ra, Dec of the transient if position is not randomly drag, see simulate_cube.py with parameter -p
same setting compared to branch master with additional functions to plot some settings
Additional component if a transient is simulated or not --> Easy way to get a false alarm rate
Old branch for same studies as in Different_Threshold_Studies
Change wavelet thresholding in k index and stationary methods
Simulations for finding usefull templates in different Repo stored here
