Geospatial Utilities

Coordinate transforms and sensor geometry calculations for geolocating image pixels. Uses WGS-84 mean Earth radius (6,371,000 m) and libm for trigonometry (no_std-compatible).

Haversine Distance

Great-circle distance between two LatLon points in meters:

$d = 2R \arcsin\sqrt{\sin^2\!\frac{\Delta\phi}{2} + \cos\phi_1 \cos\phi_2 \sin^2\!\frac{\Delta\lambda}{2}}$

let sthlm = LatLon::new(59.33, 18.07);
let gbg = LatLon::new(57.71, 11.97);
let d = haversine_distance(sthlm, gbg); // ~398 km

Ground Sample Distance

GSD (meters per pixel) from orbit and sensor parameters:

$\text{GSD} = \frac{H \cdot p}{f}$

where $H$ is altitude (m), $p$ is pixel pitch (m), and $f$ is focal length (m). The function takes pitch in micrometers and focal length in millimeters.

Parameter	Unit
altitude_m	meters
focal_length_mm	millimeters
pixel_pitch_um	micrometers

Example: 500 km altitude, 50 mm focal length, 5.5 um pixel pitch gives GSD = 55 m/px.

Swath Width

Cross-track footprint from altitude and field of view:

$W = 2H\tan\!\frac{\theta}{2}$

Parameter	Unit
altitude_m	meters
fov_deg	degrees

Pixel to LatLon

pixel_to_latlon converts image pixel coordinates to geographic coordinates using a nadir approximation. The image center is assumed to be at the sub-satellite point (nadir), and pixels are uniformly spaced at gsd meters.

Parameter	Description
px, py	Pixel coordinates
center_px, center_py	Image center in pixel coordinates
nadir	Sub-satellite point (LatLon)
gsd	Ground sample distance (m/px)

The approximation is valid for narrow fields of view where Earth curvature within the image footprint is negligible.

Coordinate Types

LatLon holds latitude (-90 to 90) and longitude (-180 to 180) in degrees. deg2rad and rad2deg convert between units.