tm_radon_forward — 2D tomography op

• 資料種類:image → image

• 呼叫:fullseye.apply(img, "tm_radon_forward", a=0.5, b=0.5)(2-D 的模型是一張圖 + 兩個純量旋鈕 a,b∈[0,1])

用法

> 該運算子的說明尚無譯文,以下照原文給出。

Forward Radon projection: treat `v` as a slice image and integrate it

along parallel rays to build a sinogram (rows = angles, cols = detector),

refit to the original HxW. `a` sets how many angles are acquired

(`n = round(H*a) clamped to 8..360 -- the sparse-view knob), b` sets the

angular span (`span = 180*min(1, 0.5+b)` deg: b < 0.5 is limited-angle CT,

b >= 0.5 -- including the default b = 0.5 -- is the full [0,180) span that

`tm_fbp_reconstruct / tm_sart_reconstruct / tm_backproject_unfiltered`

assume, so the natural chain forward -> inverse round-trips at default knobs).

Uses `skimage.transform.radon` when available, else the NumPy

rotate-and-sum Radon transform.

詳細使用指南

• gallery2d_physics_alife_3d 族使用指南

參考(範例資料・文獻)

• 範例資料目錄(下載 URL / 授權) —— 2-D 用 skimage.data(BSD/公有領域)加合成圖,3-D 給出真實資料源(Stanford/PDS 等)的下載 URL。

• 運算子來歷與參考文獻 —— 該運算子族所依據的研究/方法出處。

• 演算法的正典(作者・年份)與用途見上面的族使用指南。

可執行的範例(實際呼叫該運算子並已驗證的樣例)

• gallery2d_physics_alife_3d — py -3.11 examples/gallery2d_physics_alife_3d.py

型別可銜接的下一個運算子(可接受 image 作為輸入)

identity · gaussian · mean_box · bilateral · unsharp · median · min_filter · max_filter

同類別(tomography)

tm_fbp_reconstruct · tm_sart_reconstruct · tm_backproject_unfiltered · tm_sinogram_denoise

*Provenance: ops.py — 2D 運算子登記表。本條目由 tools/opdocs.py md 自動產生(請勿手動編輯)。*

© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.