Files
Jungfraujoch/image_analysis/indexing/FFTIndexerCPU.cpp
T
leonarski_f 6dfe065365
Build Packages / Create release (push) Successful in 16s
Build Packages / build:rugnux:aarch64 (cross) (push) Successful in 8m27s
Build Packages / build:rugnux-tgz (x86_64) (push) Successful in 9m15s
Build Packages / build:viewer-tgz:cpu (push) Successful in 10m11s
Build Packages / build:viewer-tgz:cuda (push) Successful in 12m6s
Build Packages / build:rpm (rocky8_nocuda) (push) Successful in 15m44s
Build Packages / build:rpm (rocky9_nocuda) (push) Successful in 16m1s
Build Packages / build:windows:nocuda (push) Successful in 17m29s
Build Packages / build:windows:cuda (push) Successful in 19m58s
Build Packages / HDF5 consumer tests (DIALS, XDS) (push) Successful in 24m7s
Build Packages / build:rpm (ubuntu2404_nocuda) (push) Successful in 19m8s
Build Packages / build:rugnux:windows (push) Successful in 10m58s
Build Packages / build:rpm (ubuntu2204_nocuda) (push) Successful in 20m46s
Build Packages / Generate python client (push) Successful in 53s
Build Packages / build:rpm (rocky8_sls9) (push) Successful in 20m13s
Build Packages / Build documentation (push) Successful in 1m36s
Build Packages / build:rpm (rocky9_sls9) (push) Successful in 19m57s
Build Packages / build:rpm (rocky8) (push) Successful in 18m7s
Build Packages / build:rpm (rocky9) (push) Successful in 18m54s
Build Packages / build:rpm (ubuntu2204) (push) Successful in 19m32s
Build Packages / build:rpm (ubuntu2404) (push) Successful in 17m30s
Build Packages / Unit tests (push) Successful in 1h39m2s
v1.0.0-rc.172 (#82)
* Fixed `jfjoch_broker` cancelling every data collection with a CUDA "out of memory" error after long operation: GPU memory no longer leaks with each collection.
* Rugnux scales a rotation sweep until the per-frame scales settle instead of for a fixed three rounds, and says so when they did not - merged intensities, and the space group, resolution cut and frame rejection read off them, change accordingly; `--scaling-iterations` is now the cap on that loop (default 100).
* Rugnux places every frame of a marCCD, SMV or miniCBF series at the spindle angle its own header states, so a series with missing frames, or with angles written modulo 360, is no longer read at the wrong geometry or refused.
* Every rotation run writes two diagnostic files beside its reflections: `<prefix>_detector.jpg`, the detector projection with the pixel mask and the detected beam-stop shadow drawn on it, and `<prefix>_plot.txt`, one row per image.

Reviewed-on: #82
Co-authored-by: Filip Leonarski <filip.leonarski@psi.ch>
2026-09-22 06:48:37 +02:00

136 lines
5.2 KiB
C++

// SPDX-FileCopyrightText: 2025 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
// SPDX-License-Identifier: GPL-3.0-only
#include "FFTIndexerCPU.h"
#include <cmath>
#include <algorithm>
#include <stdexcept>
#include <cassert>
#include <mutex>
#include <vector>
#include "../../common/FFTWPlannerLock.h"
static inline double dot_abs(const Coord& a, const Coord& b) {
return std::fabs(a.x * b.x + a.y * b.y + a.z * b.z);
}
FFTIndexerCPU::FFTIndexerCPU(const IndexingSettings& settings)
: FFTIndexer(settings) {
// Allocate host buffers
h_input_fft.resize(input_size, 0.0);
h_output_fft.resize(output_size);
// Validate allocations vs. expected sizes
if (h_input_fft.size() != input_size)
throw std::runtime_error("FFTWIndexer: input buffer size mismatch");
if (h_output_fft.size() != output_size)
throw std::runtime_error("FFTWIndexer: output buffer size mismatch");
const int H = static_cast<int>(histogram_size);
const int out_len = (H / 2) + 1;
int n[1] = { H };
{
std::unique_lock ul(FFTWPlannerMutex());
plan = fftwf_plan_many_dft_r2c(
1, n, nDirections,
h_input_fft.data(), nullptr, 1, H,
reinterpret_cast<fftwf_complex*>(h_output_fft.data()), nullptr, 1, out_len,
FFTW_ESTIMATE);
}
if (!plan)
throw std::runtime_error("fftw_plan_many_dft_r2c failed");
}
FFTIndexerCPU::~FFTIndexerCPU() {
std::unique_lock ul(FFTWPlannerMutex());
fftwf_destroy_plan(plan);
}
void FFTIndexerCPU::ExecuteFFT(const std::vector<Coord> &coord, size_t nspots) {
// Build histograms: one per direction
const int H = static_cast<int>(histogram_size);
const int D = nDirections;
const int out_len = (H / 2) + 1;
std::fill(h_input_fft.begin(), h_input_fft.end(), 0.0);
for (int d = 0; d < D; ++d) {
float* hist = h_input_fft.data() + static_cast<size_t>(d) * H;
for (size_t i = 0; i < nspots; i++) {
const auto& r = coord[i];
double dot = dot_abs(direction_vectors[d], r);
long long bin = static_cast<long long>(dot / histogram_spacing);
if (bin >= 0 && bin < H) {
hist[bin] += 1.0;
}
}
}
// Plan and execute batched R2C FFT with FFTW
fftwf_execute(plan);
// Post-process: pick the peak past min_length_A by PROMINENCE above a local background.
// The projected histogram has a broad low-frequency ENVELOPE (spots cluster near the
// origin) whose magnitude can exceed the true lattice peaks; a plain argmax|spec| then
// returns a short envelope vector (~10A) on weak/pink-beam frames and the real axes are
// lost. Subtracting a running-mean background of half-width BG_HALF bins removes that
// smooth envelope (it cancels to ~0) while sharp lattice peaks - fundamentals AND
// harmonics - keep their height. The prominence is also reported as the magnitude so
// FilterFFTResults ranks directions by real-peak strength, not by envelope.
const double len_coeff = 2.0 * static_cast<double>(max_length_A) / static_cast<double>(H);
// Background half-window ~15 A (in length, so it is independent of the histogram sizing);
// wide enough to span the envelope yet narrow enough not to smooth real peaks. Validated
// optimum on serial-still jet data (de-novo indexing improved markedly vs FFBIDX); a second dataset unchanged.
constexpr double BG_HALF_WIDTH_A = 15.0;
const int bg_half = std::max(1, static_cast<int>(std::lround(BG_HALF_WIDTH_A / len_coeff)));
std::vector<double> mag(out_len), pref(out_len + 1);
for (int d = 0; d < D; ++d) {
const auto* spec = h_output_fft.data() + static_cast<size_t>(d) * out_len;
pref[0] = 0.0;
for (int j = 0; j < out_len; ++j) {
mag[j] = std::hypot(spec[j][0], spec[j][1]);
pref[j + 1] = pref[j] + mag[j];
}
double best_prom = 0.0;
double best_len = -1.0;
for (int j = 0; j < out_len; ++j) {
double len = len_coeff * static_cast<double>(j);
if (len <= static_cast<double>(min_length_A)) continue;
// Slide the background window inward at the ends rather than truncating it. A peak within
// bg_half bins of either end - which is where the LONGEST cells sit, the last usable bin
// being max_length_A itself - otherwise gets its background from a one-sided window, and
// the prominence it is judged on is biased by however much the spectrum slopes there.
int lo = j - bg_half, hi = j + bg_half + 1;
if (lo < 0) { hi = std::min(out_len, hi - lo); lo = 0; }
if (hi > out_len) { lo = std::max(0, lo - (hi - out_len)); hi = out_len; }
const double background = (pref[hi] - pref[lo]) / static_cast<double>(hi - lo);
const double prominence = mag[j] - background;
if (prominence > best_prom) {
best_prom = prominence;
best_len = len;
}
}
result_fft[d] = FFTResult{
.magnitude = static_cast<float>(best_prom),
.direction = d,
.length = static_cast<float>(best_len)
};
}
}