checkerboard op• Data kinds: image2d → keypoints
• Call: import fullseye as fs; fs.ledger.find_checkerboard_corners(image, pattern_size, sigma=1.5, refine_window=None) (to call the implementation directly, import calib_lm; calib_lm.find_checkerboard_corners(image, pattern_size, sigma=1.5, refine_window=None); from the registry, opscalib.get("find_checkerboard_corners"))
• GPU: none (CPU reference implementation only)
Detects the inner corners of a checkerboard image to sub-pixel accuracy and returns them in the board's row-major order (keypoints).
> The detailed description below is the original text — the summary and the headings are translated.
Detect the inner corners of a checkerboard with pattern_size = (cols, rows) inner corners, refine them to sub-pixel accuracy and return them as (row, col) in the board's row-major order (pairs with checkerboard_points).
手順: (1) ガウスの 2 階微分(尺度 sigma)で鞍点の強さ −det(Hessian) を出し、局所最大を候補に
(2) 各候補を Förstner の勾配直交の式(OpenCV の cornerSubPix と同じ原理: 窓の中の勾配 g が角へ向かう
ベクトルと直交 → Σ g gᵀ (p − q) = 0)で反復して詰める(Förstner & Gülch 1987)
(3) 最寄りの近傍から格子の 2 方向を決め、局所の平行四辺形で予測しながら格子を伸ばす
(4) 埋まった格子が cols x rows(または転置)でなければ拒否。
順の約束: 列の方向 = 長さ cols の軸。2 つの軸の向きは「画像で右手系(板を表から見る)」と「最初の角が画像の左上に最も近い」で決める。★対称な板の 180 度の回転は区別しない —— 校正には効かない(板の点の集合がその回転で自分に写るので、姿勢が 180 度回るだけ)。cols = rows の板では 90 度も区別しない。
返す (N, 2): (row, col)、N = cols·rows。
Raises ValueError: 画像が 2 次元でない・非有限、角の数が合わない・格子が組めない(板が見えない・ぼけすぎ・一部が画像の外)、引数の範囲外。
• camera_calibration family guide
• 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_calibration_truth_map — py -3.11 examples/poc_calibration_truth_map.py
keypoints as input)image_to_world_plane · camera_calibration · camera_calibration_lm · stereo_calibration_lm · vector_to_rigid · vector_to_similarity · hom_vector_to_proj_hom_mat2d
checkerboard)checkerboard_points · sim_checkerboard
*Provenance: calib_lm.py — CALIB 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.