walkterrain op• Data kinds: depth → depth
• Call: import fullseye as fs; fs.ledger.ground_surface(grid, cell: 'float' = 0.05, radius: 'float' = 0.4) (to call the implementation directly, import terrain; terrain.ground_surface(grid, cell: 'float' = 0.05, radius: 'float' = 0.4); from the registry, opsdrive.get("ground_surface"))
• GPU: none (CPU reference implementation only)
Smooth walkable-ground envelope by grey-opening (min-filter then max-filter).
> The detailed description below is the original text — the summary and the headings are translated.
A morphological opening with a structuring element radius wide follows gentle undulation and erases anything narrower that sticks *up*, which makes it a good ground model for rough / curved terrain. Honest limit: a flat structuring element cannot sit under a steep ramp at the grid's up-slope border, so on a strong planar slope the envelope lags by ~radius*slope there and detect_obstacles with ground='opening' can report a phantom obstacle in the last radius/cell cells — use the default ground='plane' for planar/ramp ground. nan cells are filled first.
空の入力(grid が 0 要素)は形を保った空の配列を返す(理由: 観測の無い地図には地面の包絡も無く、形だけを保つ)。
• Sample-data catalog (download URLs / licences) — 2-D uses skimage.data (BSD/public domain) plus synthetic images; 3-D lists download URLs for real data sources (Stanford, PDS, …).
• Operator provenance and references — the sources of the research/methods this op family came from.
• The canonical algorithm (author, year) and its uses are named in the family usage guide above.
• walking_ops_tour — py -3.11 examples/walking_ops_tour.py
depth as input)spin_from_marker_sequence · carla_transform_matrix · carla_pose_to_world · carla_camera_pose · carla_xy_yaw · carla_scene_load · town_chain · osm_parse
walkterrain)elevation_map · fill_gaps · elevation_traversability · foothold_score · ground_plane · detect_obstacles · fuse_elevation · roughness_map
*Provenance: terrain.py — DRIVE operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.