sign_image — DRIVE roadjp op

• Data kinds: none → rgba (an op determined by its arguments alone — it takes no image or data input)

• Call: import fullseye as fs; fs.ledger.sign_image(kind: 'str', *, size_px: 'int' = 128, value=None, font_path=None, side: 'float | None' = None) -> 'np.ndarray' (to call the implementation directly, import roadjp; roadjp.sign_image(kind: 'str', *, size_px: 'int' = 128, value=None, font_path=None, side: 'float | None' = None) -> 'np.ndarray'; from the registry, opsdrive.get("sign_image"))

Usage

Returns a picture of the road sign as RGBA float (H, W, 4) ∈ [0,1] (outside the plate = alpha 0). Width = size_px; the height follows the shape's aspect ratio.

Shapes and colours (standard colour schemes, symbols simplified):

• stop: inverted triangle, red background, white "Tomare" (Japanese for "stop") + a small "STOP" below. slow: inverted triangle, white background, red border, blue "徐行" (Japanese for "slow down") + "SLOW".

• speed_limit: circle, white background, red border, blue digits (value, default 40). no_entry: circle, red background, white horizontal bar.

• no_parking: circle, blue background, red border, red diagonal (top-left → bottom-right). one_way: wide blue rectangle with a white arrow (pointing right).

• crosswalk: blue pentagon (house shape) with a white pedestrian and crosswalk stripes.

• caution_crossing / caution_signal / caution_children: yellow diamond, black border, black symbol (railway crossing = 2 barrier arms and ×, traffic light = 3 circles (red, yellow, blue), schoolchildren = 2 human figures).

Text is rasterised with PIL (same font search as annotate; can be fixed with font_path). If the font has no glyph, ValueError (it does not silently output tofu boxes). side is the same as in sign_params (only affects the size ratio).

References (sample data, literature)

• 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.

Runnable examples (verified samples that actually call this op)

• gallery_drive_japan_roads — py -3.11 examples/gallery_drive_japan_roads.py

Ops the type connects to (they accept rgba as input)

spin_from_marker_sequence · plate_mesh_from_image · carla_transform_matrix · carla_pose_to_world · carla_camera_pose · carla_xy_yaw · carla_scene_load · town_chain

Same category (roadjp)

sign_params · plate_mesh_from_image · sign_mesh · add_sign · signal_jp_mesh · add_signal_jp


*Provenance: roadjp.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.