可以将文章内容翻译成中文,广告屏蔽插件可能会导致该功能失效(如失效，请关闭广告屏蔽插件后再试):

问题:

I have seen similar questions but so far none that fit my needs (at least to the extent I understand them) so I will apologize in advance if this has already been answered. I am also kind of an R novice.

I have a data frame containing two sets of Lat/Lon in each row. The actual data contains hundreds of rows and many columns of related data. I am plotting the points for both sets of Lat/Lon and want to draw lines connecting each pair as a separate line. Here is an example of what the results should look like.

[![enter image description here][1]][1]

Here is a simplified example of the data. There will be duplicate Event and Location values.

Event_lat   Event_lon   Event   Location    Location_latitude   Location_longitude
40.791151   -124.054008 704832643   60005   40.790961   -124.1825609
38.900882   -122.660353 704653051   60009   38.873889   -122.709722
38.921488   -122.600049 704681147   60011   38.85111099 -122.593333
38.921488   -122.600049 704681147   60011   38.85111099 -122.593333
39.141877   -123.044724 706777142   60012   39.22794396 -123.064722
38.928113   -122.611386 708644013   60016   38.98950003 -122.7695828
39.02361    -122.72195  708582623   60016   38.98950003 -122.7695828
38.87586    -122.842684 708336092   60016   38.98950003 -122.7695828
39.239926   -123.145497 709020144   60017   39.24138798 -123.2163878
39.3307 -123.221674 708875205   60017   39.24138798 -123.2163878

Here is a simplified sample of the code to map the points:

library(leaflet)
myData <-read.csv("Book1.csv",header=TRUE, sep=",")
leaflet()%>%
  addTiles() %>%
  addCircles(myData,lng = myData$lsr_lon,lat = myData$lsr_lat, radius=20, color = "red",group = "events") %>% 
  addCircles(myData,lng = myData$site_longitude,lat = myData$site_latitude, radius=20, color = "blue",group = 'Locations')

回答1:

I myself am just starting with geographic use of R and leaflet, so this probably is not the most efficient way to approach this problem.. Nevertheless it is working fine for me... Feedback is always welcome!

result

sample data

df <- read.table( text = "Event_lat   Event_lon   Event   Location    Location_latitude   Location_longitude
40.791151   -124.054008 704832643   60005   40.790961   -124.1825609
38.900882   -122.660353 704653051   60009   38.873889   -122.709722
38.921488   -122.600049 704681147   60011   38.85111099 -122.593333
38.921488   -122.600049 704681147   60011   38.85111099 -122.593333
39.141877   -123.044724 706777142   60012   39.22794396 -123.064722
38.928113   -122.611386 708644013   60016   38.98950003 -122.7695828
39.02361    -122.72195  708582623   60016   38.98950003 -122.7695828
38.87586    -122.842684 708336092   60016   38.98950003 -122.7695828
39.239926   -123.145497 709020144   60017   39.24138798 -123.2163878
39.3307 -123.221674 708875205   60017   39.24138798 -123.2163878", header = TRUE)

code

I want to create a spatiallines-object, that I can draw in leaflet with addPolylines.

First, i want to create a data.frame with lat/lon only, with a unique id per combination of event-location.

library(tidyverse)
#craete a column with unique id's per event-location combination
df <- df %>% mutate( id = row_number() )
#create a temporaty df with events
events.df <- df %>% 
  select( id, Event_lat, Event_lon) %>% 
  rename( latitude = Event_lat, longitude = Event_lon)
#create a temporaty df with locations
locations.df <- df %>% 
  select( id, Location_latitude, Location_longitude) %>%
  rename( latitude = Location_latitude, longitude = Location_longitude)
#merge the two temp.df's together
df.sp <- bind_rows( events.df, locations.df )

#    id latitude longitude
# 1   1 40.79115 -124.0540
# 2   2 38.90088 -122.6604
# 3   3 38.92149 -122.6000
# 4   4 38.92149 -122.6000
# 5   5 39.14188 -123.0447
# 6   6 38.92811 -122.6114
# 7   7 39.02361 -122.7220
# 8   8 38.87586 -122.8427
# 9   9 39.23993 -123.1455
# 10 10 39.33070 -123.2217
# 11  1 40.79096 -124.1826
# 12  2 38.87389 -122.7097
# 13  3 38.85111 -122.5933
# 14  4 38.85111 -122.5933
# 15  5 39.22794 -123.0647
# 16  6 38.98950 -122.7696
# 17  7 38.98950 -122.7696
# 18  8 38.98950 -122.7696
# 19  9 39.24139 -123.2164
# 20 10 39.24139 -123.2164

Now to create the spatialLines-object

library(maptools)
library(sp)
#make df.sp a spatialdataframe
coordinates( df.sp ) <- c( "longitude", "latitude" )

#create a list per id
id.list <- sp::split( df.sp, df.sp[["id"]] )

#initialisation of counter
id <- 1

#for each id, create a line that connects all points with that id
for ( i in id.list ) {
  event.lines <- SpatialLines( list( Lines( Line( i[1]@coords ), ID = id ) ),
                               proj4string = CRS( "+init=epsg:4326" ) )
  if ( id == 1 ) {
    sp_lines  <- event.lines
  } else {
    sp_lines  <- spRbind( sp_lines, event.lines )
  }
  id <- id + 1
}

check.

head(sp_lines,1)

# An object of class "SpatialLines"
# Slot "lines":
#   [[1]]
# An object of class "Lines"
# Slot "Lines":
#   [[1]]
# An object of class "Line"
# Slot "coords":
#   longitude latitude
# [1,] -124.0540 40.79115
# [2,] -124.1826 40.79096
# 
# Slot "ID":
#   [1] "1"
# 
# Slot "bbox":
#   min        max
# x -124.18256 -124.05401
# y   40.79096   40.79115
# 
# Slot "proj4string":
#   CRS arguments:
#   +init=epsg:4326 +proj=longlat +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0

Now, draw the points and the polylines in leaflet

library(leaflet)
leaflet()%>%
  addTiles() %>%
  addCircles(df,lng = df$Event_lon, lat = df$Event_lat, radius=20, color = "red", group = "events") %>% 
  addCircles(df,lng = df$Location_longitude, lat = df$Location_latitude, radius=20, color = "blue", group = 'Locations') %>%
  addPolylines( data = sp_lines )

回答2:

I propose a solution using library(sf) and library(data.table), where sf has superseeded sp, and data.table is used for efficient reshaping of data.

I am using the data provided by Wimpel in their solution.

Reshaping

Creating an sf object is fairly straight forward. We need to create a LINESTRING for each row of your data (as an sfg object), then convert to sf

library(sf)
library(data.table)

setDT(df)   

## create an 'id' / index value. This assumes each row of your data is a separate line. 
df[, idx := .I]

## create an `sfc` column (where each row is an `sfg` object)
sf <- df[
    , {
        geometry <- sf::st_linestring(x = matrix(c(Event_lon, Event_lat, Location_longitude, Location_latitude), ncol = 2, byrow = T))
        geometry <- sf::st_sfc(geometry)
        geometry <- sf::st_sf(geometry = geometry)
    }
    , by = idx
]

## convert to sf
sf <- sf::st_as_sf(sf)

Plotting

With this sf object you can now plot in leaflet (using similar code to Wimpel)

library(leaflet)

leaflet() %>%
    addTiles() %>%
    addPolylines(data = sf) %>%
    addCircles(data = df, lng = ~Event_lon, lat = ~Event_lat, radius=20, color = "red", group = "events") %>% 
    addCircles(data = df, lng = ~Location_longitude, lat = ~Location_latitude, radius=20, color = "blue", group = 'Locations')