Connecting spatiallines or SpatialPoints based on minimum distance using R

Question

Does anybody know of a way to create a single polygon (that will eventually be converted to an Owin) from either a spatialpoints object or a spatiallines object?

Here is an example of each:

Here's some very cut down code for this:

test.verts <- list()
test.verts$x <- c(518, 594, 421, 286, 104, 129, 315)
test.verts$y <- c(273, 608, 1202, 1201, 620, 285, 135)

test.owin <- spatstat::owin(poly = test.verts)

cltest <- create.lines(test.owin, nlines=20, width = 5,spacing=NULL,angle=0)

x0list <- c(cltest$x0)
meanx0 <- lapply(x0list, mean)
y0list <- c(cltest$y0)
meany0 <- lapply(y0list, mean)

x1list <- c(cltest$x1)
meanx1 <- lapply(x1list, mean)
y1list <- c(cltest$y1)
meany1 <- lapply(y1list, mean)

pspowin <- owin(xrange=c(min(unlist(meanx0)),max(unlist(meanx1))), 
                yrange=c(min(unlist(meany0)),max(unlist(meany0))))

psplines1 <- as.psp(cltest, window = pspowin)

coords <- psplines1$ends
nc <- nrow(coords)
out <- vector(mode="list", length=nc)

for (i in seq(along=out)) {
  lLine <-Line(cbind(c(coords[i,1], coords[i,3]), c(coords[i,2],
                                                    coords[i,4])))
  out[[i]] <- Lines(list(lLine), ID=i)
}

aix.lines <- SpatialLines(out)

plot(aix.lines) #lines needing to be connected

sflines <- st_as_sf(aix.lines)

pt <- st_cast(sflines, "POINT") #points needing to be connected; can also convert to SpatialPoints from sp

plot(pt)

Answer 1

For the example you've shown a convex hull is the easiest way to enclose all points in a polygon. For a sp object you could use the convex hull algorithm from rGeos. I'm not sure if rGeos is compatible with sf. If any of your desired polygons has a concave vertex you'll need a routine that traces the points in a specified order.