如何用R语言sf包将指定经纬度序列转换为sf多边形数据框

Hey there! I'll walk you through two straightforward ways to turn your coordinate data frame into the sf polygon object you need. Let's jump right in.

Prerequisite #

First, make sure you have the sf package installed and loaded:

install.packages("sf") # Run this only if you haven't installed it yet
library(sf)

Your original data frame is already set, but let's restate it for clarity:

df <- data.frame(
  order = 1:5,
  lon = c(119.4,119.4,119.4,119.5,119.5),
  lat = c(-5.192,-5.192,-5.187,-5.187,-5.191)
)

Method 1: Direct Polygon Construction (Most Explicit) #

This method gives you full control over the coordinate structure, which is great for understanding how sf polygons work under the hood.

1. Ensure coordinates are in the correct sequence:
   Even though your order column already sorts the points, it's good practice to explicitly sort to avoid any accidental misordering:

   df_sorted <- df[order(df$order), ]
   

2. Create a closed coordinate matrix:
   SF requires polygons to be closed (start and end at the same point). We'll add the first coordinate pair to the end of our matrix to meet this requirement:

   coords <- rbind(df_sorted[, c("lon", "lat")], df_sorted[1, c("lon", "lat")])
   

3. Build the polygon and sf data frame:
   Convert the coordinate matrix into a polygon object, then wrap it into an sf feature collection with the correct CRS (WGS84, EPSG:4326):

   # Create a single polygon (note the nested list structure required by st_polygon)
   poly <- st_polygon(list(as.matrix(coords)))
   # Convert to sf data frame with CRS set
   sf_poly <- st_sfc(poly, crs = 4326) %>% st_sf()
   

Method 2: Casting from Points (More Concise) #

If you prefer a clean pipeline-style approach, you can convert the points to an sf object first, then cast them up to a polygon:

sf_poly <- df %>%
  st_as_sf(coords = c("lon", "lat"), crs = 4326) %>% # Convert raw data to sf points
  arrange(order) %>% # Ensure points follow your intended sequence
  st_combine() %>% # Combine points into a single multi-point object
  st_cast("LINESTRING") %>% # Cast to a connected line string
  st_cast("POLYGON") %>% # Cast line string to polygon (sf auto-closes it if needed)
  st_sf() # Wrap into a proper sf data frame

Check the Result #

If you print sf_poly, you'll get output exactly matching your desired format:

## Simple feature collection with 1 feature and 0 fields
## geometry type:  POLYGON
## dimension:      XY
## bbox:           xmin: 119.4 ymin: -5.192 xmax: 119.5 ymax: -5.187
## epsg (SRID):    4326
## proj4string:    +proj=longlat +datum=WGS84 +no_defs
##                         geometry
## 1 POLYGON ((119.4 -5.192, 119...

内容的提问来源于stack exchange，提问作者rasyidstat