Tutorials#

Two end-to-end notebooks. Both use openly available data; no NASA Earthdata or Vantor credentials, no API keys.

1. ASTER Mt. Rainier

Stereo from a single ASTER L1A scene. The gentlest end-to-end flow:

  • aster2asp to extract nadir + back-looking views

  • parallel_stereo (raw, then mapprojected for cleaner output)

  • point2dem to grid the result

  • asp_plot for diagnostics + PDF report

Output: ~30 m DEM of Mt. Rainier.

Tutorial 1 — ASTER Mt. Rainier
2. WorldView-3 UCSD

Stereo from a high-resolution commercial-style WV3 pair (SpaceNet CORE3D, openly hosted on AWS). The full recipe:

  • Stereo geometry analysis with StereoGeometryPlotter

  • COP-DEM clip from AWS Open Data

  • bundle_adjust on full images

  • mapproject to the COP-DEM grid

  • parallel_stereo on the mapprojected pair

  • asp_plot --pc_align for ICESat-2 alignment

Output: ~1 m DEM of UCSD campus.

Tutorial 2 — WorldView-3 UCSD

Suggested order#

If you’ve never run ASP before, do the ASTER tutorial first. It has fewer parameters, and a coarser resolution.

The WorldView tutorial then adds bundle adjustment, cropping, and a finer resolution, higher accuracy result.

Beyond the tutorials#

The asp_plot example notebooks cover jitter correction, planetary missions (LRO NAC, Mars MOC NA, CTX), no-mapprojection variants, and scene selection.