symbol_capture op• データ種: table → image2d
• 呼び出し: import fullseye as fs; fs.ledger.render_symbol(symbol, module_px=4.0, *, quiet=None, bar_height=None, rotate_deg=0.0, tilt_deg=0.0, tilt_axis_deg=0.0, distance_ratio=10.0, offset_px=(0.0, 0.0), ink_gain=0.0, dot=None, dark=0.05, light=0.85, margin_modules=2.0, supersample=4, return_geometry=False) (実装を直接呼ぶなら import symrender; symrender.render_symbol(symbol, module_px=4.0, *, quiet=None, bar_height=None, rotate_deg=0.0, tilt_deg=0.0, tilt_axis_deg=0.0, distance_ratio=10.0, offset_px=(0.0, 0.0), ink_gain=0.0, dot=None, dark=0.05, light=0.85, margin_modules=2.0, supersample=4, return_geometry=False)、台帳から引くなら opssymbol.get("render_symbol"))
• GPU: なし(CPU 参照実装のみ)
シンボル(1-D の幅か 2-D のモジュール格子)を、平面の反射率画像として描く(image2d)。
Render a symbol as the reflectance image a pinhole camera sees. module_px is the module (X dimension) size in pixels at the symbol centre; rotate_deg turns the symbol in its plane (counter-clockwise on screen), tilt_deg tilts the plane about an in-plane axis at tilt_axis_deg (perspective; the camera sits distance_ratio symbol sizes away, so 10 means mild keystone). ink_gain (modules) widens dark areas (bar width gain, > 0) or narrows them (< 0); dot draws 2-D dark modules as discs of that diameter (modules). Reflectances are dark (ink) and light (substrate); each pixel is the mean of supersample² point samples (area coverage). The canvas holds the symbol, its quiet zone (quiet modules; default 4 for QR and raw grids, 2 for the table from dm_encode unless the table says otherwise; for 1-D the table's own quiet zones and quiet is the margin above and below the bars) and a further margin_modules of substrate.
return_geometry=True returns a table instead: image, homography (plane [u, v, 1] in modules → [col, row, w]), plane_size (modules), module_px.
空の入力(モジュール格子が 0 行)は ValueError。拒否(ValueError): 表に modules も edges も無い、module_px <= 0.5、dark >= light、反射率が [0, 1] の外、|tilt_deg| >= 80、|ink_gain| > 0.9、supersample が 1..16 の外、キャンバスが 4000 画素を超える。
• サンプルデータ カタログ(DL URL / ライセンス) — 2-D は skimage.data(BSD/public)+ 合成、3-D は実データ源(Stanford/PDS 等)の DL URL。
• 演算子の来歴・参考文献 — この op 族の元になった研究/手法の出典。
• アルゴリズムの正典(著者・年)と用途は上記ファミリ使い方ガイドに記載。
• poc_symbol_package_1d — py -3.11 examples/poc_symbol_package_1d.py
• poc_symbol_part_dpm — py -3.11 examples/poc_symbol_part_dpm.py
image2d を入力に取れる)barcode_scan_profiles · read_barcode1d · qr_locate · qr_sample · qr_read · dm_locate · dm_sample · dm_read
symbol_capture)*Provenance: symrender.py — SYMBOL operator registry. この per-op ノートは tools/opdocs.py md が自動生成(手編集しない)。*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.