cutting op• Data kinds: rgb → table
• Call: import fullseye as fs; fs.ledger.knife_edge_track(image, board_row: 'float | None' = None, win: 'int' = 6, reject_px: 'float' = 1.0, method: 'str' = 'coverage') -> 'dict' (to call the implementation directly, import cutting; cutting.knife_edge_track(image, board_row: 'float | None' = None, win: 'int' = 6, reject_px: 'float' = 1.0, method: 'str' = 'coverage') -> 'dict'; from the registry, opsdrive.get("knife_edge_track"))
Extracts the cutting-edge line, the spine and the tip from the face-view image (rule-based only).
Procedure: per column, pick the row where the vertical luminance step goes "dark → bright" (blade → wall) → exclude food columns (saturation rule) and 3 px around them → refine to sub-pixel by the coverage method (window widened to match the blur estimate) → total-least-squares line (same formulation as measure.fit_line) → drop residuals > reject_px and refit. The upper boundary (spine and the oblique line at the tip) is picked the same way; the continuous part on the right that departs from the spine line = the oblique line, and its intersection with the edge line = the tip. For columns hidden by the food (occluded_cols), the edge is the interpolation of a single line fitted from both sides.
method: "coverage" (default) or "gradient" (centroid of the derivative, for comparison). Returns: angle_deg (counter-clockwise with x right and z up, atan2), line (total least squares of row = cy + (col − cx)·dy/dx), slope, offset, tip_rc (tip; None if not found), n_points, rms_px, cols_used, occluded_cols. ValueError if no blade is visible.
• 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.
• poc_food_cutting_measure — py -3.11 examples/poc_food_cutting_measure.py
table as input)course_layout · course_occupancy · course_contains · world_build · world_camera · world_move · lidar_scan · rss_longitudinal_same
cutting)cutting_scene · cutting_face_render · cutting_edge_render · cutting_episode_synth · cutting_wrist_mjcf · cut_depth_from_side · slice_thickness_profile · cut_surface_roughness
*Provenance: cutting.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.