Get values from parallel features

This function finds features in a 'target' network (target_net) that are parallel and close to features in a 'source' network (source_net) and imputes missing values in a specified column of the target_net based on the median value from the nearby parallel features in the source_net.

Usage

get_parallel_values(
  target_net,
  source_net,
  column = "maxspeed",
  buffer_dist = 10,
  angle_threshold = 20,
  value_pattern = " mph",
  value_replacement = "",
  add_suffix = " mph"
)

Arguments

target_net

An sf object representing the network where values need imputation (e.g., cycle network). Must contain a unique osm_id column.

source_net

An sf object representing the network to source values from (e.g., drive network). Must contain a unique osm_id column.

column

The name of the column (as a string) in both networks to check for NAs in target_net and source values from source_net. Defaults to "maxspeed". This column should ideally be numeric or coercible to numeric after removing units.

buffer_dist

The buffer distance (in the units of the CRS) around target_net features to search for nearby source_net features. Defaults to 10.

angle_threshold

The maximum allowed absolute difference in bearing (degrees) between features in target_net and source_net for them to be considered parallel. Defaults to 20.

value_pattern

Optional regex pattern to remove from the column values in source_net before converting to numeric. Defaults to " mph". Set to NULL to skip removal.

value_replacement

Replacement string for value_pattern. Defaults to "".

add_suffix

Optional suffix to add back to the imputed numeric value before assigning it back to the column. Defaults to " mph". Set to NULL to skip adding suffix.

Value

An sf object, the target_net with the specified column potentially updated with imputed values from source_net.

Details

It's a generalisation of the find_nearby_speeds function previously used to impute missing maxspeed values in a cycle network based on nearby roads.

Examples

# Assuming cycle_net_f and drive_net_f are loaded sf objects from the package

# Impute missing 'maxspeed' in cycle_net_f using drive_net_f
cycle_net_updated_speed = get_parallel_values(
  target_net = cycle_net_f,
  source_net = drive_net_f,
  column = "maxspeed",
  buffer_dist = 10,
  angle_threshold = 20
)
#> Warning: st_point_on_surface assumes attributes are constant over geometries
#> Warning: st_point_on_surface may not give correct results for longitude/latitude data

# Example imputing a hypothetical 'width' column (assuming it exists and needs imputation)
# cycle_net_f$width = as.character(cycle_net_f$width) # Ensure character if adding suffix
# cycle_net_f$width[sample(1:nrow(cycle_net_f), 5)] = NA # Add some NAs for example
# cycle_net_updated_width = get_parallel_values(
#   target_net = cycle_net_f,
#   source_net = drive_net_f, # Using drive_net just for example structure
#   column = "width",
#   buffer_dist = 5,
#   angle_threshold = 15,
#   value_pattern = NULL, # Assuming width is already numeric or has no suffix
#   add_suffix = NULL
# )

print(paste("NA maxspeed before:", sum(is.na(cycle_net_f$maxspeed))))
#> [1] "NA maxspeed before: 28"
print(paste("NA maxspeed after:", sum(is.na(cycle_net_updated_speed$maxspeed))))
#> [1] "NA maxspeed after: 20"