Name List

The name list is a Python dictionary containing the necessary parameters for configuring the tracking process. Some parameters are mandatory for complete tracking, while others are optional, allowing for customization and improved performance. You can view the full list of parameters by using the command pyfortracc.default_parameters or by opening the pyfortracc/default_parameters.py file.

Mandatory Parameters

  • input_path: Path to the input files.

  • output_path: Path for saving the output tracking files.

  • thresholds: A Python list of intensity thresholds used in the segmentation process.

  • min_cluster_size: A list specifying the minimum cluster size (in number of points) for each threshold during clustering.

  • operator: String representing the comparison operator used to separate groups in the segmentation process (e.g., ‘>=’, ‘<=’, ‘==’).

  • timestamp_pattern: File name pattern for reading and timestamp conversion.

  • delta_time: Time interval for each file or the desired interval between frames in the track (in minutes).

Example of a Name List with Mandatory Parameters

# Empty list
name_list = {}
# Path to the input data
name_list['input_path'] = 'input/'
# Path to save output data
name_list['output_path'] = 'output/'
# Thresholds in dBZ for segmentation
name_list['thresholds'] = [20, 30, 40]
# Minimum points per cluster for each threshold
name_list['min_cluster_size'] = [10, 5, 3]
# Comparison operator: '>=', '<=', or '=='
name_list['operator'] = '>='
# Pattern for timestamp extraction from filenames
name_list['timestamp_pattern'] = '%Y%m%d_%H%M%S.nc'
# Time interval in minutes
name_list['delta_time'] = 12

Optional Parameters

These parameters allow you to include geospatial information in the tracking process. If they are not set, the algorithm will proceed without geospatial data.

  • lon_min: Minimum longitude in degrees.

  • lon_max: Maximum longitude in degrees.

  • lat_min: Minimum latitude in degrees.

  • lat_max: Maximum latitude in degrees.

Example with Optional Parameters

# Minimum longitude
name_list['lon_min'] = -62.1475
# Maximum longitude
name_list['lon_max'] = -57.8461
# Minimum latitude
name_list['lat_min'] = -5.3048
# Maximum latitude
name_list['lat_max'] = -0.9912

Additional Parameters

These parameters provide advanced configuration options for pyForTraCC, enabling users to customize the tracking process further. They are not mandatory but can be used to fine-tune the algorithm for specific use cases.

  • mean_dbz: If True, the mean reflectivity is used for tracking. Default is False.

  • cluster_method: Clustering method to use. Options are ‘ndimage’ or ‘dbscan’. Default is ‘ndimage’.

  • eps: Epsilon distance for clustering when using the ‘dbscan’ method. Default is 1.

  • delta_tolerance: Maximum time difference (in minutes) allowed between two files for tracking. Default is 0.

  • num_prev_skip: Number of previous files to skip during tracking. Default is 0.

  • edges: Whether to use domain edges for cluster linking. Options are True or False. Default is False.

  • n_jobs: Number of parallel jobs to run. Default is -1 (uses all available cores).

  • min_overlap: Minimum overlap (in pixels) between clusters to consider them the same. Default is 10.

  • convex_hull: If True, the convex hull is used to calculate cluster geometry. Default is False.

  • preserv_split: If True, split lifetime events are preserved for NEW/SPLIT events. Default is False.

  • spl_correction: Enables vector correction for split events. Default is False.

  • mrg_correction: Enables vector correction for merge events. Default is False.

  • inc_correction: Enables vector correction for inner cells. Default is False.

  • opt_correction: Enables optical flow vector correction. Default is False.

  • opt_mtd: Optical flow method to use. Options are ‘farneback’ or ‘lucas-kanade’. Default is ‘lucas-kanade’.

  • elp_correction: Enables vector correction using ellipse fitting. Default is False.

  • calc_dir: If True, calculates the direction of cluster movement. Default is False. (coming soon)

  • calc_speed: If True, calculates the speed of cluster movement. Default is False. (coming soon)

  • speed_units: Units for speed calculation. Default is ‘m/s’. (coming soon)

  • epsg: EPSG code for spatial projection. Default is 4326. (coming soon)

Example with Additional Parameters

In the example below, the parameters indicate that the algorithm was configured to track using the opt_correction method, which enables the optical-flow method and uses the opt_mtd Farneback to enhance the vector components. Another parameter added is min_overlap, which specifies that the overlap percentage was changed to 15%.

# Enable optical flow correction
name_list['opt_correction'] = True
# Use Farneback method for optical flow
name_list['opt_mtd'] = 'farneback'
# Set minimum overlap for cluster matching
name_list['min_overlap'] = 15