leonarski_fandClaude Opus 5
a6be35ccdb
The profile is the MEAN of each bin, so a few strong reflections landing in a bin lift it exactly as a smooth powder ring does. That is the wrong quantity whenever the profile is wanted as a background rather than as a measurement of what is in the bin - the ice score being the case in point, where reading a plain profile INVERTED the metric: over 37 rotation crystals the two highest-scoring crystals had no ice at all. The adaptive spot finder already computes the right thing, a sigma-clipped per-resolution-ring background, as a byproduct of its own threshold. Where it runs, the ice score uses that. Where it does not - --no-adaptive-spots, --azint-only, and anything reading the profile the broker wrote - there was no way to get it. This adds one: azim_int_settings.sigma_clip (rugnux --azim-sigma-clip), 0 = off, minimum 2 because a tighter clip rejects a large part of a clean Gaussian bin and biases the estimate low rather than removing outliers. Two clip passes follow the plain one, matching the finder's recipe - the first pass's standard deviation is itself inflated by the peaks being removed, so one pass leaves a threshold that is still too generous. A bin with fewer than eight pixels is left alone: at the detector edge and behind the beam stop there is no spread to clip on. Both engines do it. On the GPU the accept range is computed by a small kernel and stays resident, so a clip pass is one more read of the same pixels and no round trip; the two accumulation kernels take the range as a pointer that is null on the plain pass. Measured on a JUNGFRAU rotation dataset, non-adaptive path: azimuthal integration 0.02 -> 0.06 ms per image, exactly the 3x the extra passes predict, against a 0.34 ms per-image total. Note what the result IS: the smooth background under the peaks, not the bin mean. It should not be switched on where a ring's integrated intensity is wanted - the powder-ring geometry fit reads ring peaks, and those are what a clip is designed to remove. Off by default, so nothing changes unless it is asked for. Not exposed over the REST API - that needs the generated model regenerated, which is a separate step. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
37 lines
1.6 KiB
C++
37 lines
1.6 KiB
C++
// SPDX-FileCopyrightText: 2025 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
|
|
// SPDX-License-Identifier: GPL-3.0-only
|
|
|
|
#pragma once
|
|
|
|
#include "../../common/AzimuthalIntegrationMapping.h"
|
|
#include "../../common/AzimuthalIntegrationProfile.h"
|
|
#include "../image_preprocessing/ImagePreprocessorBuffer.h"
|
|
|
|
class AzIntEngine {
|
|
protected:
|
|
const AzimuthalIntegrationMapping& integration;
|
|
const uint16_t azint_bins;
|
|
const size_t npixel;
|
|
std::vector<float> azint_sum;
|
|
std::vector<float> azint_sum2;
|
|
std::vector<uint32_t> azint_count;
|
|
|
|
// Sigma clipping (AzimuthalIntegrationSettings::SigmaClip; 0 = off). Two clip passes follow the
|
|
// plain one - the same one-plain-plus-two recipe the adaptive spot finder uses, where the second
|
|
// matters because the first pass's standard deviation is itself inflated by the peaks being
|
|
// removed, so one pass alone leaves a threshold that is still too generous.
|
|
static constexpr int CLIP_PASSES = 2;
|
|
const float clip_nsigma;
|
|
std::vector<float> clip_lo;
|
|
std::vector<float> clip_hi;
|
|
|
|
// Per-bin accept range from the accumulators of the pass just finished. A bin with too few pixels
|
|
// to have a meaningful spread is left unclipped rather than guessed at.
|
|
void UpdateClipLimits();
|
|
[[nodiscard]] int PassCount() const { return clip_nsigma > 0.0f ? 1 + CLIP_PASSES : 1; }
|
|
public:
|
|
AzIntEngine(const AzimuthalIntegrationMapping& integration);
|
|
virtual ~AzIntEngine() = default;
|
|
virtual void Run(const ImagePreprocessorBuffer &image, AzimuthalIntegrationProfile &profile) = 0;
|
|
};
|