typed op• 데이터 종류: beatcube → image
• 호출: fullseye.apply(img, "tb_range_doppler_map", a=0.5, b=0.5)(2-D 는 이미지 1 장 + 스칼라 노브 2 개 a,b∈[0,1] 모델)
*그림 없음: 이 연산자는 beatcube를 입력으로 받습니다. 이미지에서 시작하는 Studio 프로그램으로는 그 타입에 닿지 않으므로 아래의 실행 가능한 예제를 보세요.*
비트 큐브(beat cube)의 2차원 FFT -> `(n_doppler, n_range)` 크기(magnitude) 맵.
> 아래 상세 설명은 원문입니다 —— 요약과 제목은 번역되어 있습니다.
Fast time transforms to range (last axis, not shifted: bin `j` is
`j * c*f_s/(2*S*N_s)` metres, and a physical range is always positive so
the whole `[0, f_s)` band is used). Slow time transforms to velocity
(middle axis, `fftshift`ed so the map is centred on zero velocity: bin
`i is (i - N_c//2) * lambda/(2*N_c*T_c)` metres per second, positive =
receding).
The antenna axis is collapsed by *combine*: `"incoherent"` (default) is the
mean of the magnitudes, which is angle independent and therefore the
right default for detection; `"coherent"` is the magnitude of the mean,
i.e. a beam pointed at boresight, which attenuates an off-boresight target on
purpose. `antenna=k uses element k` alone. For a single-element cube
all three agree exactly.
`normalize=True divides by N_c * N_s`, so a bin-centred target of
amplitude `a peaks at exactly a` (measured: 1.0 for a unit target,
absolute error 0.0). The default `False` keeps the raw FFT magnitude.
No window is applied — compose :func:fmcw_window_apply first if you want
one. The output is a plain 2-D float64 array, so every 2-D operator in
Fullseye (threshold, morphology, labelling, blob measurement — the pieces a
CFAR detector is made of) applies to it directly.
Raises `ValueError`: a real-valued cube (it would put a mirror ghost of
every target at a fabricated range), fewer than 2 chirps or 2 samples, an
out-of-range *antenna* index, an unknown *combine*, a cube over the element
cap, an FFT that overflows to NaN, or NaN/Inf on the way in.
Typed bridge of the rangedoppler op `range_doppler_map 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.
• 샘플 데이터 카탈로그(DL URL / 라이선스) —— 2-D 는 skimage.data(BSD/public)+ 합성, 3-D 는 실데이터 소스(Stanford/PDS 등)의 DL URL.
• 연산자의 내력·참고문헌 —— 이 연산자 족의 바탕이 된 연구/기법의 출처.
• (아직 없음)
image 를 입력으로 받는 것)identity · gaussian · mean_box · bilateral · unsharp · median · min_filter · max_filter
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 연산자 레지스트리. 이 op 노트는 tools/opdocs.py md 가 자동 생성합니다(직접 편집하지 마세요).*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.