Adding PSF content and structure

environment.yml

@@ -7,5 +7,6 @@ dependencies:
   - matplotlib
   - numpy
   - scipy
+  - galsim
   - pip:
     - jupyter-book

shearbook/_toc.yml

@@ -6,6 +6,17 @@
     - file: intro/about
     - file: intro/run_code
 
+- part: Point Spread Function
+  chapters:
+    - file: psf/psf-intro
+      sections:
+        - file: psf/psf-origins-nb
+    - file: psf/psf-modelling
+      sections:
+        - file: psf/non-parametric-model
+        - file: psf/parametric-model
+        - file: psf/psf-models-ref
+
 - part: Shape Measurement
   chapters:
     - file: moments/moments-intro

shearbook/psf/non-parametric-model.md

@@ -0,0 +1,15 @@
+# Non-parametric
+
+```{warning}
+In progress!
+```
+
+Describe the PSF modelling problem. Use stars to build a model and try to recover the PSF at galaxy positions.
+
+Describe the main idea of a non-parametric model of the PSF.
+
+Give an example, PSFEx (and MCCD maybe?).
+
+_(Note) Should I give a more mathematical description of how one of the models (PSFEx for example) is build?_
+
+_Maybe I can add some figures from the MCCD paper. Showing some of the eigenPSFs learned. Also, some observed CFIS star and the estimated PSF model._

shearbook/psf/parametric-model.md

@@ -0,0 +1,13 @@
+# Parametric
+
+```{warning}
+In progress!
+```
+
+Describe the PSF modelling problem. Characterise the model as much as possible and try to fit the reduced number of parameters with observed stars. Then use the model to recover PSFs at galaxy positions.
+
+Describe the main idea of a parametric model of the PSF.
+
+Give an example, TinyTim.
+
+_(Note) Should I give an example like of how the PSF is generated following an optic model? _

shearbook/psf/psf-intro.md

@@ -0,0 +1,8 @@
+# Introduction to the PSF
+
+In this section we will..
+
+**Contents**
+
+```{tableofcontents}
+```

shearbook/psf/psf-modelling.md

@@ -0,0 +1,8 @@
+# PSF Modelling
+
+In this section we will go through the two main families of Point Spread Function modelling methods.
+
+**Contents**
+
+```{tableofcontents}
+```

shearbook/psf/psf-models-ref.md

@@ -0,0 +1,5 @@
+# References
+
+```{bibliography} ../_bibliography/z_psf_models.bib
+:all:
+```

shrbk/plot_psfs.py

@@ -0,0 +1,36 @@
+import numpy as np
+import matplotlib.pyplot as plt
+from mpl_toolkits.axes_grid1 import make_axes_locatable
+
+def plot_psf(psf_img, cmap='gist_stern'):
+    fig = plt.figure(figsize=(18,4))
+
+    ax = fig.add_subplot(131)
+    im = ax.imshow(psf_img, interpolation='none', origin='lower', cmap=cmap)
+    ax.set_xlabel('x [pixels]')
+    ax.set_ylabel('y [pixels]')
+    divider = make_axes_locatable(ax)
+    cax = divider.append_axes('right', size='5%', pad=0.05)
+    fig.colorbar(im, cax=cax, orientation='vertical')
+    ax.set_title('PSF')
+
+
+    ax = fig.add_subplot(132)
+    im = ax.imshow(np.log(abs(psf_img)), interpolation='none', origin='lower', cmap=cmap)
+    ax.set_xlabel('x [pixels]')
+    ax.set_ylabel('y [pixels]')
+    divider = make_axes_locatable(ax)
+    cax = divider.append_axes('right', size='5%', pad=0.05)
+    fig.colorbar(im, cax=cax, orientation='vertical')
+    ax.set_title('log(PSF)')
+
+    ax = fig.add_subplot(133)
+    im = ax.imshow(abs(np.fft.fftshift(np.fft.fft2(psf_img))), interpolation='none', origin='lower', cmap=cmap)
+    ax.set_xlabel('x [freq]')
+    ax.set_ylabel('y [freq]')
+    divider = make_axes_locatable(ax)
+    cax = divider.append_axes('right', size='5%', pad=0.05)
+    fig.colorbar(im, cax=cax, orientation='vertical')
+    ax.set_title('abs(FFT(PSF))')
+
+    plt.show()