R 对多边形进行分类并对其加权

嗨，我有这个数据集：

> head(sp_csv_data)
Longitude   Latitude Radius Site_Type
1 -177.87567 -24.715167     10       MIG
2  -83.21360  14.401800      1       OBS
3  -82.59392   9.589192      1       NES
4  -82.41060   9.492750      1   NES;BRE
5  -81.17555   7.196750      5       OBS
6  -80.95770   8.852700      1       NES

经度和纬度是站点中心的坐标。场地是圆，半径列是场地的半径。最后一列是场地类型

我需要将这些.csv信息转换为光栅。我能够画出圆圈并将其光栅化。现在我需要根据站点类型对它们进行分类。到目前为止，该层的外观如下所示：

> print(sp_robinson)
Simple feature collection with 46519 features and 1 field
geometry type:  POLYGON
dimension:      XY
bbox:           xmin: -16675160 ymin: -5378567 xmax: 16655730 ymax: 6089138
CRS:            +proj=robin +lon_0=180 +x_0=0 +y_0=0 +ellps=WGS84 +datum=WGS84 +units=m +no_defs
First 10 features:
   layer                       geometry
1     69 POLYGON ((11889163 6089138,...
2     69 POLYGON ((11927764 6089138,...
3     69 POLYGON ((11966365 6089138,...
4     73 POLYGON ((12622585 6089138,...
5     73 POLYGON ((12661186 6089138,...
6     73 POLYGON ((12699787 6089138,...
7     73 POLYGON ((12738389 6089138,...
8     73 POLYGON ((12776990 6089138,...
9     73 POLYGON ((12815591 6089138,...
10    73 POLYGON ((12854192 6089138,...

你知道我该如何根据场地类型对圆圈进行分类吗

这是我到目前为止使用的代码：

lon <- c(-177.87567, -83.2136, -82.59392, -82.4106, -81.17555, -80.9577)
lat <- c(-24.715167, 14.4018, 9.589192, 9.49275, 7.19675, 8.8527)
radius <- c(10, 1, 1, 1, 5, 1)
site <- c("MIG", "OBS", "NES", "NES;BRE", "OBS", "NES")
sp_csv_data <- data.frame(lon, lat, radius, site)

####Defining Datum and Coordinates system (best to do first to avoid errors later)
rob_pacific <- "+proj=robin +lon_0=180 +x_0=0 +y_0=0 +ellps=WGS84 +datum=WGS84 +units=m +no_defs" 
longlat <- "+proj=longlat +datum=WGS84 +ellps=WGS84 +towgs84=0,0,0"

####Creating the nodes####
    
nodes <- sp_csv_data %>%
  st_buffer(dist = sp_csv_data$Radius)

####Converting to raster####
 
# Creating a empty raster at 0.5° resolution (you can increase the resolution to get a better border precision)
rs <- raster(ncol = 360*2, nrow = 180*2) 
rs[] <- 1:ncell(rs)
crs(rs) <- CRS(longlat)

##Convering to raster
sp_raster <- rasterize(nodes, rs)

# Resampling to make sure that it's in the same resolution as sampling area
sp_raster <- resample(sp_raster, rs, resample = "ngb")

#converting into an sf spatial polygon dataframe
sp_raster <- as(sp_raster, "SpatialPolygonsDataFrame")
species_sp <- spTransform(sp_raster, CRS(longlat))

# Define a long & slim polygon that overlaps the meridian line & set its CRS to match that of world
polygon <- st_polygon(x = list(rbind(c(-0.0001, 90),
                                     c(0, 90),
                                     c(0, -90),
                                     c(-0.0001, -90),
                                     c(-0.0001, 90)))) %>%
  st_sfc() %>%
  st_set_crs(longlat)

# Transform the species distribution polygon object to a Pacific-centred projection polygon object
sp_robinson <- species_sp %>% 
  st_as_sf() %>% 
  st_difference(polygon) %>% 
  st_transform(crs = rob_pacific)

# There is a line in the middle of Antarctica. This is because we have split the map after reprojection. We need to fix this:
bbox1 <-  st_bbox(sp_robinson)
bbox1[c(1,3)]  <-  c(-1e-5,1e-5)
polygon1 <- st_as_sfc(bbox1)
crosses1 <- sp_robinson %>%
  st_intersects(polygon1) %>%
  sapply(length) %>%
  as.logical %>%
  which
# Adding buffer 0
sp_robinson[crosses1, ] %<>%
  st_buffer(0)

---

lon我向您展示了如何使您的示例更具可复制性。请相应地编辑您的问题，谢谢您的帮助和建议！我真的很感激，我对这个很陌生。。。我希望这是更清楚，更可复制！我想说的是，在我的回答中，我展示了如何使代码具有可复制性，因为您可以复制和粘贴代码，然后运行它。因此，如果你从这开始，你也可以显示你在哪里卡住了。你正在做很多事情。我的建议是首先在lon/lat光栅上进行光栅化。如果你有，下一件事就是担心你想要使用的不寻常的crs。谢谢！我现在明白多了！正如你所看到的，我能够光栅化我的圆圈，现在我所需要的就是对它们进行分类！我现在明白了，我在重做很多无用的步骤，但我认为现在它更一般，更容易。再次感谢！