You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
The current implementation of the reproject_geom function simply maps geometry coordinates from one projection to another. For high distortion projections, this can result in the reprojected geometry not faithfully representing the distortion. For example, compare the reprojected geometry to the reprojected data when mapping from a sinusoidal projection to geographic WGS84 coordinates:
The blue line is the result of simply reprojecting the geometry coordinates. The green line more accurately represents the projection distortion by adding additional vertices to the geometry coordinate list before reprojection.
Describe the solution you'd like
An option to insert additional vertices into the geometry coordinate list in order to more faithfully represent the distortion would be good. Even better would be an option to specify the "maximum error" that is acceptable in the reprojected geometry.
Describe alternatives you've considered
I've put together an example of one approach here that accepts an error "tolerance" (maximum error). One flaw is that it is brute force in the number of vertices added to the geometry prior to reprojection, which results in excess computations. A more elegant solution for densifying the geometry based on projection distortion is desirable, perhaps based on something like Tissot's indicatrix, provided the complexity does not outweigh the current brute force approach.
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
The current implementation of the
reproject_geom
function simply maps geometry coordinates from one projection to another. For high distortion projections, this can result in the reprojected geometry not faithfully representing the distortion. For example, compare the reprojected geometry to the reprojected data when mapping from a sinusoidal projection to geographic WGS84 coordinates:The blue line is the result of simply reprojecting the geometry coordinates. The green line more accurately represents the projection distortion by adding additional vertices to the geometry coordinate list before reprojection.
Describe the solution you'd like
An option to insert additional vertices into the geometry coordinate list in order to more faithfully represent the distortion would be good. Even better would be an option to specify the "maximum error" that is acceptable in the reprojected geometry.
Describe alternatives you've considered
I've put together an example of one approach here that accepts an error "tolerance" (maximum error). One flaw is that it is brute force in the number of vertices added to the geometry prior to reprojection, which results in excess computations. A more elegant solution for densifying the geometry based on projection distortion is desirable, perhaps based on something like Tissot's indicatrix, provided the complexity does not outweigh the current brute force approach.
The text was updated successfully, but these errors were encountered: