This folder contains scripts for analysing camera data and calibration data using the methods described in the SPECTACLE paper (https://doi.org/10.1364/OE.27.019075). These scripts are used to create plots, tables, etc. which provide insight into the performance and characteristics of the camera.
This README is only intended to give a very broad overview of the results that can be obtained using these scripts. For further documentation, please refer to that included in the scripts themselves.
The following scripts were used to generate the figures from the SPECTACLE paper:
- Fig. 2: linearity_characterise_multiple.py, based on output from linearity_raw.py and linearity_jpeg.py
- Fig. 3: linearity_plot_response_multiple.py
- Fig. 4: jpeg_sRGB_comparison_plot_multiple.py, based on output from jpeg_sRGB_gamma_free.py and jpeg_sRGB_gamma_fixed.py
- Fig. 5: plot_RGBG2_multiple.py
- Fig. 6: iso_normalisation_multiple.py, based on output from iso_normalisation.py
- Fig. 7: gain_characterise_multiple.py, based on output from gain.py
- Fig. 8: gain_characterise_multiple.py, based on output from gain.py
- Fig. 9: flatfield_compare_maps.py, based on output from flatfield.py
- Fig. 10: spectral_response_plot_multiple.py, based on output from spectral_response_monochromator.py and spectral_response_ispex_sun.py
- Fig. 11: This script was generated using a highly hard-coded version of spectral_response_ispex_sun.py. However, it is more easily replicated using the more general spectral_response_plot_multiple.py script.
Please note that for the paper, some parameters such as axis labels were manually edited.
plot_RGBG2_multiple.py may be used to plot Gaussed maps of any number of data stacks, similar to Fig. 7 in the paper.
Pearson r maps may be generated using linearity_raw.py (for RAW data) and linearity_jpeg.py (for JPEG data). Histograms of the resulting Pearson r distributions may be plotted using linearity_characterise.py (for a single camera) or linearity_characterise_multiple.py (for multiple cameras) to create plots similar to Fig. 2 in the paper.
The response of the central pixels in a camera may be analysed using linearity_plot_response.py (for a single camera) or linearity_plot_response_multiple.py (for multiple cameras) to create plots similar to Fig. 3 in the paper.
The JPEG response of a camera may be analysed using jpeg_sRGB_gamma_free.py (fitting an sRGB response with a free gamma value to each pixel) and jpeg_sRGB_gamma_fixed.py (comparing them to sRGB models with fixed gamma values). The results from this may be plotted using jpeg_sRGB_characterise_model.py (free gamma), jpeg_sRGB_comparison_plot.py (fixed gamma), or jpeg_sRGB_comparison_plot_multiple.py (free and fixed gamma, multiple cameras) to create plots similar to Fig. 4 in the paper.
Bias maps may be analysed using bias_characterise.py to create plots similar to Fig. 5 in the paper.
Read noise maps, in ADU, may be analysed using readnoise_characterise_ADU.py to create plots similar to Fig. 5 in the paper. Another script, readnoise_characterise_normalised.py, does the same but normalises data according to ISO speed first.
Furthermore, the relation between read noise and ISO speed may be analysed using readnoise_iso_relation.py to create a plot of read noise at each ISO speed.
The dark current in a camera may be analysed using using dark_characterise_ADU.py to create histograms and Gaussed maps of dark current per pixel in normalised ADU/s. If a gain map has been created, dark_characterise_electrons.py may be used to create similar figures in units of electrons/s.
The ISO speed normalisation may be analysed using iso_normalisation.py (single camera) or iso_normalisation_multiple.py (multiple cameras) to generate plots similar to Fig. 6 in the paper.
Gain maps may be analysed using gain_characterise.py (single camera) or gain_characterise_multiple.py to create plots similar to Figs. 7 and 8 in the paper.
Flat-field data may be analysed using flatfield_characterise_data.py to create plots similar to Fig. 5 in the paper and plot the SNR of the data.
Flat-field models may be compared using flatfield_compare_model_parameters.py (comparing model parameters) or flatfield_compare_maps.py (comparing fitted flat-field maps). The latter may also be used to create plots similar to Fig. 9 in the paper.
Spectral response curves may be compared at each wavelength using spectral_response_compare_curves.py.
Monochromator data may be analysed using spectral_response_monochromator_single_spectrum.py (for a spectrum taken with a single setting, e.g. filter/grating) and spectral_response_monochromator_plot_outputs.py (to plot intermediary results from the calibration process).
Spectral response curves may be plotted using spectral_response_plot_multiple.py to create plots similar to Fig. 10 in the paper.
The SRFs may be compared to the CIE xyz colour-matching functions, and the camera's colour space may be compared to that of the human eye and sRGB in xy space using xyz_matrix_plot.py. To compare multiple cameras to each other, one can use xyz_matrix_plot_multiple.py.