typed opqimage → imagefullseye.apply(img, "tb_monogenic_orientation", a=0.5, b=0.5) (2-D は 1 画像 + 2 スカラつまみ a,b∈[0,1] のモデル)
図は合成の入力 128×128 で実際に走らせた出力。左が入力、右が出力。点群は上から見た散布(明るさ = z)、1-D 列は折れ線、体積は z 方向の最大値投影、動画は中央フレーム、複素画像は振幅、絵にならない返り値は値そのもの。
出力は viridis 風の疑似カラー(暗い紫 = 小、黄 = 大)。距離・位相・向き・深度のような「量の場」を読むため。
つまみ a は出力を変えない(実測: 0.1 / 0.5 / 0.9 で同一)。
つまみ b は出力を変えない(実測: 0.1 / 0.5 / 0.9 で同一)。
段階(前置きの op → この op。左から順):

別の画像でも(合成シーン / 写真 / 硬貨。上段が入力、下段がその出力。つまみは既定):

Local orientation atan2(R2, R1) of a monogenic signal. → (H, W).
Radians in [0, pi): an orientation is defined modulo pi (a grating at
10 degrees and one at 190 degrees are the same grating), and the value is
folded into that range rather than left in (-pi, pi] where the same
structure would read as two different numbers on either side of a contrast
reversal. display=True maps it to [0, 1].
Continuous, not quantised — the angle is read directly from two filters,
for any angle, where a steerable bank with K orientations interpolates
between its K. Measured against eight grid-exact grating orientations the
error is at most 3.6e-15 rad, including the obliques. (Whether that
buys anything downstream is a separate question, and the measured answer is
mostly no — see :func:riesz_displacement.)
Where it is undefined, and the mask is not the one you expect. The
orientation dies where the Riesz vector dies, which is at every
even-symmetric point — local phase 0 or pi, the crest of a bright or dark
line — and the amplitude is at full strength there. Measured on a 45-degree
grating, the worst orientation error over the whole frame is 0.2764 rad, at a
pixel where |R| = 6.8e-16 and :func:monogenic_amplitude reads
1.0000. So masking on the amplitude does not protect you; mask on
hypot(q[..., 1], q[..., 2]), the Riesz magnitude. With that mask the
error over the same eight orientations is at most 3.6e-15 rad.
Where the Riesz vector is exactly zero, atan2(0, 0) = 0 is returned —
a value, not a measurement.
Raises ValueError: the input is not a valid quaternion field, or its
k component is non-zero; display is not a bool.
Typed bridge of the quat op monogenic_orientation into the 2-D evolution registry: the same implementation, called under the op(v, a, b) convention. This op has no tunable parameter; a and b are unused.
下のプログラムは実際に走ることを確かめてある(図と同じ入力)。Studio のヘルプではこのブロックがボタンになり、その場で読み込んで実行できる。
img_to_monogenic 0.50 0.50
tb_monogenic_orientation 0.50 0.50
次の例は元の台帳 op monogenic_orientation を呼ぶもの。この橋渡し op は同じ実装を fn(v, a, b) 規約に合わせただけなので、挙動はそのまま当てはまる(呼び出し形だけ違う)。
py -3.11 examples/perf_monogenic_metrics.pypy -3.11 examples/quaternion_monogenic.pyimage を入力に取れる)identity · gaussian · mean_box · mean_box_wh · bilateral · kuwahara · unsharp · median
typed)tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_project_points · tb_render_point_depth · tb_points_render · tb_statistical_outlier_removal · tb_radius_outlier_removal
Provenance: ops.py — 2D 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.