Skip to contents

Identify "N" gap boundaries where a NONE is found between animal images

Source: R/cam_obtain_n_gap_class.R
cam_obtain_n_gap_class.Rd

Flags boundaries where an animal image is followed (after one or more rows of NONE) by another animal image within a short time window. These should be split into separate series because the intervening NONE indicates the animal left the FOV, even if the total animal-to-animal image gap is small.

Usage

cam_obtain_n_gap_class(
  cons_main_report,
  grouping = c("project", "project_id", "location", "location_id"),
  datetime_col = "image_date_time",
  species_col = "species_common_name",
  none_label = "NONE",
  max_bridge_secs = 120,
  require_same_species = TRUE
)

Arguments

cons_main_report

Main report with consolidated tags (e.g., from cam_consolidate_tags()), requiring: image_date_time, image_id, species_common_name, and camera identity columns.

grouping

Character vector of camera identity columns. Only those present are used. Default: c("project","project_id","location","location_id").

datetime_col

Name of the timestamp column. Default "image_date_time".

species_col

Name of the species column. Default "species_common_name".

none_label

Label denoting a NONE image. Default "NONE".

max_bridge_secs

Numeric, maximum animal-to-animal elapsed seconds to still consider a bridged series. Default 120.

require_same_species

Logical, require same species before/after the NONE block? Default TRUE.

Value

A tibble with one row per detected boundary: grouping columns, image_id, image_date_time, species_common_name, gap_class = "N". If none are found, returns an empty tibble and warns.

See also

cam_calc_time_by_series() which accepts the output of this function via its n_gap_df parameter to force series splits at NONE-bridged boundaries.

Author

Marcus Becker