ball_detect — DRIVE balltrack op

• Data kinds: image2d → table

• Call: import fullseye as fs; fs.ledger.ball_detect(image, *, mode: 'str' = 'bright', thresh=None, color=None, color_tol: 'float' = 0.25, radius_range=(1.5, 80.0), max_candidates: 'int' = 8) -> 'list' (to call the implementation directly, import balltrack; balltrack.ball_detect(image, *, mode: 'str' = 'bright', thresh=None, color=None, color_tol: 'float' = 0.25, radius_range=(1.5, 80.0), max_candidates: 'int' = 8) -> 'list'; from the registry, opsdrive.get("ball_detect"))

Usage

Extracts ball candidates from an image: threshold → connected components → brightness-weighted subpixel centroid, equivalent radius, fill fraction.

mode = "bright" (bright blobs, thresh default = mean + 2σ) / "dark" (dark blobs) / "color" (RGB distance to color (3,) < color_tol) / "chroma" (distance of brightness-normalised colour < color_tol: robust to shading). Return value = [{"col", "row", "radius", "area", "fill", "score"}, …] (descending score; fill = area / area of the circumscribed circle, as ball-likeness). Blobs whose radius is outside radius_range are discarded.

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)

• poc_ball_bounce — py -3.11 examples/poc_ball_bounce.py

• poc_driving_longitudinal — py -3.11 examples/poc_driving_longitudinal.py

• poc_driving_school — py -3.11 examples/poc_driving_school.py

• poc_driving_weather — py -3.11 examples/poc_driving_weather.py

• poc_kendama — py -3.11 examples/poc_kendama.py

• poc_table_tennis_bounce — py -3.11 examples/poc_table_tennis_bounce.py

• poc_table_tennis_spin — py -3.11 examples/poc_table_tennis_spin.py

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

course_layout · course_occupancy · course_contains · world_build · world_camera · world_move · lidar_scan · rss_longitudinal_same

Same category (balltrack)

ball_track · kalman_ca · triangulate_dlt · track_triangulate · bounce_detect · marker_direction · spin_from_markers · spin_from_marker_sequence

*Provenance: balltrack.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.