Build Packages / build:windows:nocuda (push) Successful in 20m4s
Build Packages / Unit tests (push) Skipped
Build Packages / build:viewer-tgz:cpu (push) Successful in 16m5s
Build Packages / build:viewer-tgz:cuda (push) Successful in 17m26s
Build Packages / build:rpm (rocky8_nocuda) (push) Successful in 27m46s
Build Packages / build:rpm (rocky9_nocuda) (push) Successful in 20m17s
Build Packages / build:rpm (ubuntu2204_nocuda) (push) Successful in 26m13s
Build Packages / build:rpm (ubuntu2404_nocuda) (push) Successful in 23m17s
Build Packages / build:rpm (rocky8_sls9) (push) Successful in 28m11s
Build Packages / build:rpm (rocky9_sls9) (push) Successful in 19m30s
Build Packages / build:rpm (rocky8) (push) Successful in 24m34s
Build Packages / build:rpm (rocky9) (push) Successful in 21m30s
Build Packages / build:rpm (ubuntu2204) (push) Successful in 23m33s
Build Packages / build:rpm (ubuntu2404) (push) Successful in 20m18s
Build Packages / DIALS test (push) Successful in 18m23s
Build Packages / XDS test (durin plugin) (push) Successful in 11m30s
Build Packages / XDS test (JFJoch plugin) (push) Successful in 10m16s
Build Packages / XDS test (neggia plugin) (push) Successful in 8m2s
Build Packages / Generate python client (push) Successful in 49s
Build Packages / Build documentation (push) Successful in 1m21s
Build Packages / Create release (push) Skipped
Build Packages / build:windows:cuda (push) Successful in 29m45s
This is an UNSTABLE release. It includes many experimental features, as well as many AI generated fixes. We recommend using rc.152 for production use. * **rugnux: significantly better quality of results, and faster.** A large rework of integration, scaling, merging, geometry refinement and space-group determination, together with measurements the program previously made no attempt at - the direct beam before indexing, the beam stop, the goniometer rotation scale, and the stretches of a sweep the crystal did not deliver. A rotation dataset typically gains observations at better <I/sigma> and R_meas, and every `mx` and `scale` run writes a `<prefix>_report.txt` results report modelled on XDS's `CORRECT.LP`. Many defaults moved with it: spot detection is self-calibrating, beam-stop detection and rotation geometry post-refinement are on, resolution limits default to as far as the detector reaches, and ice-ring handling engages only where the crystal is measured to have ice. * **jfjoch_viewer:** the beam-stop shadow, the detector calibration and the beam-centre measurement are reachable from "Analyze dataset"; the settings panel reports how the sample moved and how polarized the beam was; image rendering and interaction are faster. * **Performance:** bitshuffle+LZ4 images are decoded on the GPU rather than on the host, with the bitshuffle inverse fused into preprocessing so the decompressed frame is never held in device memory. * **Broker, writer, packaging and build:** image-slot lifetime and locking fixes, per-image datasets sized by the images actually written, the Debian/Ubuntu broker package renamed to `jfjoch`, and `image_analysis` compiling under MSVC again. **Breaking change to the rugnux command line:** * `--azint-only` and `--scale` are **removed**, replaced by `--mode azint` and `--mode scale`; the full pipeline is `--mode mx` and remains the default. A script passing the old flags now fails with the list of valid modes rather than silently running the wrong one. * `-t`/`--stride` is **refused on rotation data**: skipping frames cuts every reflection's rocking curve, so the combined fulls and their partiality would be measured over frames the sweep never recorded. Select a contiguous range with `-s`/`-e` instead. `--mode azint` and `--force-still` still take a stride. **Breaking changes to OpenAPI** - regenerate the client (`jfjoch-client` 1.0.0-rc.161, `frontend/src/client`) or read the affected fields as optional: * `image_scale_b` is removed from the `plot_type` enum, so a client requesting that plot now gets an error rather than a curve. * `azim_int_settings.high_q_recipA`, `spot_finding_settings.high_resolution_limit` and `spot_finding_settings.low_resolution_limit` are no longer `required`. All three mean "no limit at that end" when unset and are omitted from the response instead of carrying a placeholder value, which raises in a client generated from an rc.160-or-earlier spec. A value of 0 is still accepted and means the same thing. **Breaking changes to the stored formats** - a consumer reading these fields must treat them as optional: * The per-image image-scale B factor is no longer computed, so `/entry/MX/imageScaleBFactor` is absent from newly written HDF5 files and the corresponding key is absent from the CBOR DataMessage and END blocks. Files written by rc.160 and earlier still contain it and still open; nothing in the pipeline reads it any more. * `_reflns.jfjoch_diffrn_ISa` now carries the whole-range `1/sqrt(a*b)` that XDS's ISa denotes, and the error-model `a` and `b` are reported in XDS's convention; the strong-reflection asymptote moves to `_reflns.jfjoch_diffrn_ISa_asymptotic`. **A file written by an earlier version carries the asymptote under the plain `ISa` name.** Reviewed-on: #71 Co-authored-by: Filip Leonarski <filip.leonarski@psi.ch>
193 lines
8.1 KiB
C++
193 lines
8.1 KiB
C++
// SPDX-FileCopyrightText: 2026 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
|
|
// SPDX-License-Identifier: GPL-3.0-only
|
|
|
|
#include <algorithm>
|
|
|
|
#include <catch2/catch_all.hpp>
|
|
|
|
#include "../common/AzimuthalIntegrationMapping.h"
|
|
#include "../image_analysis/spot_finding/AdaptiveSpotFinderCPU.h"
|
|
|
|
namespace {
|
|
|
|
SpotFindingSettings AdaptiveSettings() {
|
|
SpotFindingSettings s{};
|
|
s.adaptive_threshold = true;
|
|
s.false_pixels_per_frame = 100.0f;
|
|
s.min_pix_per_spot = 1;
|
|
s.max_pix_per_spot = 50;
|
|
s.high_resolution_limit = 0.0f;
|
|
s.low_resolution_limit = 1.0e6f;
|
|
s.high_res_gap_Q_recipA = std::nullopt;
|
|
return s;
|
|
}
|
|
|
|
} // namespace
|
|
|
|
// Raw (untransformed) geometry: the mapping is built over the raw module layout, which is smaller
|
|
// than the converted image. The finder has to walk the raw image, so a spot planted at a raw pixel
|
|
// comes back at that pixel.
|
|
TEST_CASE("AdaptiveSpotFinderCPU_RawGeometry", "[AdaptiveSpotFinder]") {
|
|
DiffractionExperiment x(DetJF4M());
|
|
x.DetectorDistance_mm(80).BeamX_pxl(1030).BeamY_pxl(1080);
|
|
x.QSpacingForAzimInt_recipA(0.05).QRangeForAzimInt_recipA(0.05, 5.0);
|
|
x.GeometryTransformation(false);
|
|
|
|
PixelMask pixel_mask(x);
|
|
AzimuthalIntegrationMapping mapping(x, pixel_mask);
|
|
|
|
const size_t w = x.GetXPixelsNum();
|
|
const size_t h = x.GetYPixelsNum();
|
|
REQUIRE(w * h == mapping.GetPixelToBin().size());
|
|
REQUIRE(w * h < static_cast<size_t>(x.GetPixelsNumConv()));
|
|
|
|
ImagePreprocessorBuffer buffer(x.GetPixelsNum());
|
|
for (size_t i = 0; i < w * h; i++)
|
|
buffer[i] = 8 + static_cast<int32_t>(i % 5); // background 8..12
|
|
|
|
// A 3x3 blob on a pixel that has a ring - with all of its neighbours on one too.
|
|
const auto &pixel_to_bin = mapping.GetPixelToBin();
|
|
size_t spot_row = 0, spot_col = 0;
|
|
for (size_t row = 100; row < h - 100 && spot_row == 0; row++) {
|
|
for (size_t col = 100; col < w - 100; col++) {
|
|
bool all_binned = true;
|
|
for (int dr = -1; dr <= 1; dr++)
|
|
for (int dc = -1; dc <= 1; dc++)
|
|
all_binned &= pixel_to_bin[(row + dr) * w + col + dc] != UINT16_MAX;
|
|
if (all_binned) {
|
|
spot_row = row;
|
|
spot_col = col;
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
REQUIRE(spot_row > 0);
|
|
|
|
for (int dr = -1; dr <= 1; dr++)
|
|
for (int dc = -1; dc <= 1; dc++)
|
|
buffer[(spot_row + dr) * w + spot_col + dc] = 200;
|
|
|
|
std::vector<bool> res_mask(x.GetPixelsNum(), false);
|
|
AdaptiveSpotFinderCPU finder(mapping);
|
|
finder.SetResolutionMask(res_mask);
|
|
const auto spots = finder.Run(buffer, AdaptiveSettings());
|
|
|
|
REQUIRE(spots.size() == 1);
|
|
CHECK(std::lround(spots[0].RawCoord().x) == static_cast<long>(spot_col));
|
|
CHECK(std::lround(spots[0].RawCoord().y) == static_cast<long>(spot_row));
|
|
}
|
|
|
|
// The property the whole engine exists for: the threshold comes from the image's OWN noise, so the
|
|
// same settings behave the same way on a frame whose background is ten times higher. A frame is built
|
|
// with background spread S around a mean, one pixel planted a few S above it (must stay unfound) and
|
|
// one planted far above (must be found); then the identical frame scaled by ten must give the identical
|
|
// answer. Any threshold that does not track the background - a constant, or one that drops the sigma
|
|
// term - finds the weak pixel in the scaled frame, or loses the strong one.
|
|
TEST_CASE("AdaptiveSpotFinderCPU_ThresholdTracksBackground", "[AdaptiveSpotFinder]") {
|
|
DiffractionExperiment x(DetJF4M());
|
|
x.DetectorDistance_mm(80).BeamX_pxl(1030).BeamY_pxl(1080);
|
|
x.QSpacingForAzimInt_recipA(0.05).QRangeForAzimInt_recipA(0.05, 5.0);
|
|
x.GeometryTransformation(false);
|
|
|
|
PixelMask pixel_mask(x);
|
|
AzimuthalIntegrationMapping mapping(x, pixel_mask);
|
|
const auto &pixel_to_bin = mapping.GetPixelToBin();
|
|
|
|
const size_t w = x.GetXPixelsNum();
|
|
const size_t h = x.GetYPixelsNum();
|
|
|
|
// Two well-separated pixels that carry a ring, so both are seen by the finder.
|
|
std::vector<size_t> planted;
|
|
for (size_t row = 300; row < h - 300 && planted.size() < 2; row += 137)
|
|
for (size_t col = 300; col < w - 300; col += 149)
|
|
if (pixel_to_bin[row * w + col] != UINT16_MAX) {
|
|
planted.push_back(row * w + col);
|
|
break;
|
|
}
|
|
REQUIRE(planted.size() == 2);
|
|
|
|
// Background takes 5 evenly spaced levels one S apart, i.e. mean + 2S and sigma = sqrt(2) S. With
|
|
// ~100 expected noise pixels per frame the cut lands near mean + 4.1 sigma = mean + 5.8 S.
|
|
const auto run_at_scale = [&](int32_t scale) {
|
|
ImagePreprocessorBuffer buffer(x.GetPixelsNum());
|
|
for (size_t i = 0; i < w * h; i++)
|
|
buffer[i] = scale * (10 + static_cast<int32_t>(i % 5));
|
|
buffer[planted[0]] = scale * (10 + 5); // mean + 3 S: below the cut
|
|
buffer[planted[1]] = scale * (10 + 30); // mean + 28 S: well above it
|
|
std::vector<bool> res_mask(x.GetPixelsNum(), false);
|
|
AdaptiveSpotFinderCPU finder(mapping);
|
|
finder.SetResolutionMask(res_mask);
|
|
return finder.Run(buffer, AdaptiveSettings());
|
|
};
|
|
|
|
const auto plain = run_at_scale(1);
|
|
const auto scaled = run_at_scale(10);
|
|
|
|
REQUIRE(plain.size() == 1);
|
|
CHECK(std::lround(plain[0].RawCoord().x) == static_cast<long>(planted[1] % w));
|
|
CHECK(std::lround(plain[0].RawCoord().y) == static_cast<long>(planted[1] / w));
|
|
|
|
// Ten times the background, ten times the noise, ten times the signal - same answer.
|
|
REQUIRE(scaled.size() == plain.size());
|
|
CHECK(std::lround(scaled[0].RawCoord().x) == std::lround(plain[0].RawCoord().x));
|
|
CHECK(std::lround(scaled[0].RawCoord().y) == std::lround(plain[0].RawCoord().y));
|
|
}
|
|
|
|
// The per-ring background the finder hands out is the peak-EXCLUDED one: a handful of very bright
|
|
// pixels planted on a ring must not move it, which is the property that lets the ice score be read
|
|
// off it instead of off the plain azimuthal profile. Rings with too few pixels to be their own
|
|
// background come back as NaN rather than as a stale value from the previous frame.
|
|
TEST_CASE("AdaptiveSpotFinderCPU_RingBackgroundExcludesPeaks", "[AdaptiveSpotFinder]") {
|
|
DiffractionExperiment x(DetJF4M());
|
|
x.DetectorDistance_mm(80).BeamX_pxl(1030).BeamY_pxl(1080);
|
|
x.QSpacingForAzimInt_recipA(0.05).QRangeForAzimInt_recipA(0.05, 5.0);
|
|
x.GeometryTransformation(false);
|
|
|
|
PixelMask pixel_mask(x);
|
|
AzimuthalIntegrationMapping mapping(x, pixel_mask);
|
|
const auto &pixel_to_bin = mapping.GetPixelToBin();
|
|
|
|
const size_t w = x.GetXPixelsNum();
|
|
const size_t h = x.GetYPixelsNum();
|
|
|
|
ImagePreprocessorBuffer buffer(x.GetPixelsNum());
|
|
for (size_t i = 0; i < w * h; i++)
|
|
buffer[i] = 8 + static_cast<int32_t>(i % 5); // background 8..12
|
|
|
|
std::vector<bool> res_mask(x.GetPixelsNum(), false);
|
|
AdaptiveSpotFinderCPU finder(mapping);
|
|
finder.SetResolutionMask(res_mask);
|
|
|
|
finder.Detect(buffer, AdaptiveSettings());
|
|
const std::vector<float> clean = finder.GetRingBackground();
|
|
REQUIRE(clean.size() == mapping.GetBinNumber());
|
|
|
|
std::vector<int64_t> pixels_per_bin(mapping.GetBinNumber(), 0);
|
|
for (size_t i = 0; i < w * h; i++)
|
|
if (pixel_to_bin[i] < mapping.GetBinNumber())
|
|
pixels_per_bin[pixel_to_bin[i]]++;
|
|
|
|
// Plant 200 bright pixels spread over one well-populated ring - far more than a real spot, so a
|
|
// plain mean would move visibly.
|
|
const uint16_t ring = static_cast<uint16_t>(
|
|
std::max_element(pixels_per_bin.begin(), pixels_per_bin.end()) - pixels_per_bin.begin());
|
|
REQUIRE(pixels_per_bin[ring] > 10000);
|
|
int planted = 0;
|
|
for (size_t i = 0; i < w * h && planted < 200; i++)
|
|
if (pixel_to_bin[i] == ring) {
|
|
buffer[i] = 100000;
|
|
planted++;
|
|
}
|
|
REQUIRE(planted == 200);
|
|
|
|
finder.Detect(buffer, AdaptiveSettings());
|
|
const std::vector<float> spiked = finder.GetRingBackground();
|
|
|
|
REQUIRE(std::isfinite(clean[ring]));
|
|
REQUIRE(std::isfinite(spiked[ring]));
|
|
CHECK(spiked[ring] == Catch::Approx(clean[ring]).epsilon(0.01));
|
|
|
|
for (size_t b = 0; b < clean.size(); b++)
|
|
CHECK(std::isfinite(clean[b]) == (pixels_per_bin[b] >= 40));
|
|
}
|