-
-
Save pigeonflight/a4b5e9939d8f189f2340444220ccc2f7 to your computer and use it in GitHub Desktop.
| import numpy as np | |
| def in_bounds(bounds,point): | |
| top_left = bounds['top_left'] | |
| bottom_right = bounds['bottom_right'] | |
| north = top_left['north'] | |
| west = top_left['west'] | |
| south = bottom_right['north'] | |
| east = bottom_right['west'] | |
| lon = point[0] | |
| lat = point[1] | |
| in_bounds = np.logical_and(np.logical_and( | |
| lat >= west, # filter the latitude | |
| lat <= east), # that are within the boundaries | |
| np.logical_and( | |
| lon <= north, # filter the longititude | |
| lon >= south)) # that are within the boundaries | |
| return in_bounds | |
| if __name__ == '__main__': | |
| top_left = {'north':40.859311, 'west':-108.525174} | |
| bottom_right = {'north':36.625119, 'west':-99.898083} | |
| bounds= { | |
| 'top_left':top_left, | |
| 'bottom_right':bottom_right | |
| } | |
| point = [39.733283, -106.670038] #correct | |
| print in_bounds(bounds,point) |
If you use Geo software, it's more common to order the coordinates as lon/lat, as this matches using an x/y axis (x-axis would be the equator/longitude). There are also examples of using a reverse z/y/x scheme, with z being the zoom level. But that's mostly just used for some display libraries.
As Vincent said, longitude is west/east from -180 to +180 degrees. Latitude is south/north from -90 to +90 degrees. Though if you want to display points on a map, those typically use Web Mercator projection, which can only deal with -85.05 to +85.05 degrees, so filtering your inputs might make sense.
If you use bounding boxes you might also have to deal with boxes crossing the +/-180 degrees line. It's easiest to just split those boxes into two distinct ones and figure out all points in each of them and than combine the results.
You reversed latitude and longitude.
point[0] is latitude and go with north and south comparison.
point[1] is longitude and go with west and east comparison.