typed op• 数据种类:matrix → matrix
• 调用:fullseye.apply(img, "tb_stat_covariance", a=0.5, b=0.5)(2-D 的模型是一张图 + 两个标量旋钮 a,b∈[0,1])
*无图: 该算子以 matrix 为输入。从图像开始的 Studio 程序无法到达该类型，请参见下方的可运行示例。*
`(N, D) 观测数据的样本协方差矩阵 → (D, D)`。
> 以下的详细说明为原文 —— 摘要与标题已翻译。
Rows are observations, columns are variables — the `(N, D)` orientation
every Fullseye point/sample API uses (note `np.cov` defaults to the
*transposed* convention). Uses the unbiased `ddof=1` estimator (divides
by `N - 1), hence the N >= 2` requirement. The diagonal holds the
per-variable sample variances; the result is symmetric positive
semi-definite by construction, so it can go straight into
:func:mat_eigh for principal axes (the covariance-ellipse workflow).
HALCON: no public tuple/matrix operator — covariance lives inside HALCON's
calibration and matching internals only.
Typed bridge of the math op `stat_covariance 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.
• 示例数据目录(下载 URL / 许可证) —— 2-D 用 skimage.data(BSD/公有领域)加合成图,3-D 给出真实数据源(Stanford/PDS 等)的下载 URL。
• 算子来历与参考文献 —— 该算子族所依据的研究/方法出处。
• (暂无)
matrix 作为输入)identity · tb_mat_pinv · tb_mat_cond · tb_stat_correlation
typed)tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_angle_3points · tb_project_points · tb_render_point_depth · tb_statistical_outlier_removal · tb_radius_outlier_removal
*Provenance: ops.py — 2D 算子登记表。本条目由 tools/opdocs.py md 自动生成(请勿手工编辑)。*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.