New read me + debug subtract background

vpicouet · Jun 2, 2021 · 44a5cb2 · 44a5cb2
1 parent 9ece11c
commit 44a5cb2
Show file tree

Hide file tree

Showing 5 changed files with 82 additions and 82 deletions.
diff --git a/README.md b/README.md
@@ -2,6 +2,11 @@
 
 pyDS9plugin is the first open source pythonic [SAOImageDS9](https://sites.google.com/cfa.harvard.edu/saoimageds9) quick look plugin.
 
+Click on the image to see the 3 minutes youtube presentation video:
+
+[<img src="https://people.lam.fr/picouet.vincent/pyds9plugin/gif/3d2d_new.mov.gif" width="80%"
+style="display:block;margin: 0 auto;">](https://www.youtube.com/watch?v=9JzcMCxo6Fk)
+
 The documentation of the extension is available [here](https://nuage.osupytheas.fr/s/Z6D64wRpDkE7wTJ).
 
 SAOImage DS9 is an astronomical imaging and data visualization application. Its 30 years of development has made it very stable and easy to use. This made it an essential tool in all fields of astronomy (observation, simulation, instrumentation). Much more profitable but pretty unsung feature, its extensibility makes it a limitless tool to interact with astronomical data. This extensibility did not generate a large collaborative and well organized effort to develop important extensions that could progressively converge towards a stable/rapid/configurable multi-extension DS9 package.
@@ -37,7 +42,7 @@ There is not as of yet a separate conda package.
 Documentation
 -------------
 
-Some documentation is availanle available at [https://people.lam.fr/picouet.vincent/pyds9plugin](https://people.lam.fr/picouet.vincent/index.html).
+Some documentation is available available at [https://people.lam.fr/picouet.vincent/pyds9plugin](https://people.lam.fr/picouet.vincent/index.html).
 
 
 Contributions
@@ -54,20 +59,6 @@ Features
 -   Command line access: The package is totally accessible via command line. Run `DS9Utils` to see all the available function and `DS9Utils function -h` to see the help of each function. All the arguments of the functions are parsed through argparse module which makes the functions not only accessible from DS9, but from terminal and python using argv argument.
 -   Multi-processing: PyDS9plugin is by essence a quicklook plug-in that is perfect to analyze and process images on the fly by changing parameters and so on. But it was important for us to make it suitable for more important pipelines as soon as you are ok with the parameters to use. To this end, most of the functions are compatible with multi-processing so that they can be run on a set of images.
 -   Multi-operability: Command line access and python import allows to operate the plugin for other pipelines. The plugin for DS9 could actually pretty easily be operated by other visualization softwares like ginga or glueviz.
-
--   Python interpreter/macros:
-
-
-<img src="https://people.lam.fr/picouet.vincent/pyds9plugin/gif/python.mov.gif" width="80%"
-style="display:block;margin: 0 auto;">
--   VTK 3D rendering:
-
-
-<img src="https://people.lam.fr/picouet.vincent/pyds9plugin/gif/3d2d_new.mov.gif" width="80%"
-style="display:block;margin: 0 auto;">
-
--   Interactive profile fitting :
-
-
-<img src="https://people.lam.fr/picouet.vincent/pyds9plugin/gif/fitinter.mov.gif" width="80%"
-style="display:block;margin: 0 auto;">
+-   Python interpreter/macros
+-   VTK 3D rendering
+-   Interactive profile fitting
diff --git a/pyds9plugin/DS9Utils.py b/pyds9plugin/DS9Utils.py
@@ -791,14 +791,24 @@ def LoadDS9QuickLookPlugin(xpapoint=None):
     return
 
 
-def ReplaceStringInFile(path, string1, string2):
+def ReplaceStringInFile(path, string1, string2,path2=None):
     """Replaces string in a txt file"""
     fin = open(path, "rt")
     data = fin.read()
     data = data.replace(string1, string2)
     fin.close()
+    if path2 is not None:
+        path = path2
+        if os.path.exists(path):
+            os.remove(path)
+        fin = open(path, "x")
+    else:
+        # try:
+        fin = open(path, "wt")
+        # except FileNotFoundError:
+        #     fin = open(path, "x")
 
-    fin = open(path, "wt")
+    # fin = open(path, "w")
     fin.write(data)
     fin.close()
     return
@@ -866,11 +876,11 @@ def setup(xpapoint=None, config=my_conf, color="cool",argv=[]):
     # verboseprint("%s - %s - %s - %s - %s - %s" % (scale, cuts, smooth, color, invert, grid))
     cuts = np.array(args.limits.split("-"), dtype=float)
     region = getregion(d, all=False, quick=True, selected=True)
-    if region is None:
-        image_area = [0, -1, 0, -1]
-    else:
-        image_area = Lims_from_region(None, coords=region)
-    Xinf, Xsup, Yinf, Ysup = image_area
+    # if region is None:
+    #     image_area = [0, -1, 0, -1]
+    # else:
+
+    # print(Xinf, Xsup, Yinf, Ysup)
     # verboseprint(Xinf, Xsup, Yinf, Ysup)
     from astropy.io import fits
     # from tqdm import tqdm
@@ -887,14 +897,25 @@ def setup(xpapoint=None, config=my_conf, color="cool",argv=[]):
         fitsimage = d.get_pyfits()
     fitsimage = fitsimage[FitsExt(fitsimage)].data
     lx, ly = fitsimage.shape[0], fitsimage.shape[1]
-    if region is None:
-        # verboseprint("No region defined, big image, taking the center.")
+    if region is not None:
+        image_area = Lims_from_region(None, coords=region)
+        Xinf, Xsup, Yinf, Ysup = image_area
+        if (Xsup<0) | (Ysup<0):
+            image_area = [int(lx / 2), int(lx / 2) + 50, int(ly / 2), int(ly / 2) + 50]
+    else:
         image_area = [int(lx / 2), int(lx / 2) + 50, int(ly / 2), int(ly / 2) + 50]
+    Xinf, Xsup, Yinf, Ysup = image_area
+    # print(image_area)
+    # if region is None:
+    #     # verboseprint("No region defined, big image, taking the center.")
+    #     image_area = [int(lx / 2), int(lx / 2) + 50, int(ly / 2), int(ly / 2) + 50]
     image = fitsimage[Yinf:Ysup, Xinf:Xsup]
     try:
+        # print('ok')
         image_ok = image[np.isfinite(image)]
         d.set("cmap %s ; scale %s ; scale limits %0.3f %0.3f " % (args.color, args.scale, np.nanpercentile(image_ok, cuts[0]), np.nanpercentile(image_ok, cuts[1])))
     except ValueError:
+        # print('error')
         d.set("cmap %s ; scale %s " % (args.color, args.scale))
 
     return
@@ -3247,6 +3268,8 @@ def SmoothCallback(value):
             def change_contour():
                 p.update_coordinates(d["points"], mesh=mesh)
                 p.update_coordinates(d["points"], mesh=mesh_c)
+                # p.update_scalars(d["points"], mesh=mesh)
+                # p.update_scalars(d["points"], mesh=mesh_c)
                 if d["Contour"]:
                     if d["smooth"] is True:
                         p.update_coordinates(mesh_c.contour().points, mesh=contours_c)
@@ -5848,11 +5871,8 @@ def background_estimation(xpapoint=None, n=2, DS9backUp=DS9_BackUp_path, Plot=Tr
     sigma, percentile, snr, npixels, dilate = np.array([sigma, percentile, snr, npixels, dilate], dtype=int)
     box1, box2 = np.array(boxs.split(","), dtype=int)
     path = globglob(args.path)
-    jobs = []
-    manager = Manager()
-    return_dict = manager.dict()
-    if len(path) > 1:
-        Plot = False
+    # if len(path) > 1:
+    #     Plot = False
     name = Parallelize(function=BackgroundEstimationPhot, parameters=[ float(sigma), bckd, rms, (filter1, filter2), (box1, box2), n, DS9_BackUp_path, snr, npixels, dilate, percentile, mask, Plot], action_to_paralize=path, number_of_thread=args.number_processors)
 
     # for filename in path:
@@ -5894,53 +5914,41 @@ def BackgroundEstimationPhot(filename, sigma, bckd, rms, filters, boxs, n=2, DS9
         "BiweightLocationBackground": BiweightLocationBackground,
     }
     functions_rms = {"StdBackgroundRMS": StdBackgroundRMS, "MADStdBackgroundRMS": MADStdBackgroundRMS, "BiweightScaleBackgroundRMS": BiweightScaleBackgroundRMS}
-    bkg = []
-    masks = [data]
-
-    for i, mask_data in enumerate(masks):
-        if mask:
-            mask_source = make_source_mask(mask_data, nsigma=snr, npixels=npixels, dilate_size=dilate_size)
-        else:
-            mask_source = np.ones(mask_data.shape, dtype=bool)
-        bkg_estimator = functions[bckd]()
-        bkgrms_estimator = functions_rms[rms]()
-        if len(masks) > 1:
-            bkg.append(
-                Background2D(
-                    mask_data, boxs, filter_size=filters, sigma_clip=SigmaClip(sigma=sigma), bkg_estimator=bkg_estimator, exclude_percentile=exclude_percentile, bkgrms_estimator=bkgrms_estimator
-                )
-            )  # ,mask=mask_source)
-        else:
-            bkg.append(
-                Background2D(
-                    mask_data,
-                    boxs,
-                    filter_size=filters,
-                    sigma_clip=SigmaClip(sigma=sigma),
-                    bkg_estimator=bkg_estimator,
-                    exclude_percentile=exclude_percentile,
-                    bkgrms_estimator=bkgrms_estimator,
-                    mask=mask_source,
-                )
-            )
-        verboseprint("Mask %i, median = %0.2f" % (i, bkg[-1].background_median))
-        verboseprint("Mask %i, rms = %0.2f" % (i, bkg[-1].background_rms_median))
-        if i == 0:
-            fitsfile.data = fitsfile.data - bkg[-1].background  # .astype('uint16')
-        else:
-            fitsfile.data[np.isfinite(mask)] = fitsfile.data[np.isfinite(mask)] - bkg[-1].background[np.isfinite(mask)]  # .astype('uint16')
-        if len(masks) == 2:
-            masks[-1][np.isfinite(masks[-1])] = fitsfile.data[np.isfinite(masks[-1])]
-    if len(masks) == 2:
-        diff = np.nanmean(fitsfile.data[np.isfinite(masks[1])]) - np.nanmean(fitsfile.data[np.isfinite(masks[0])])
-        fitsfile.data[np.isfinite(masks[1])] -= diff
-        diff = np.nanmean(np.hstack((fitsfile.data[np.isfinite(masks[0])], fitsfile.data[np.isfinite(masks[1])])))
-        fitsfile.data[np.isfinite(masks[1])] -= diff
+    # i, mask_data = 0, data
+    # for i, mask_data in enumerate(masks):
+    if mask:
+        mask_source = make_source_mask(data, nsigma=snr, npixels=npixels, dilate_size=dilate_size)
     else:
-        diff = np.nanmean(fitsfile.data[np.isfinite(masks[0])])
-    fitsfile.data[np.isfinite(masks[0])] -= diff
-    if np.isfinite(data).all():
-        fitsfile.data = fitsfile.data.astype("uint16")
+        mask_source = np.ones(data.shape, dtype=bool)
+    bkg_estimator = functions[bckd]()
+    bkgrms_estimator = functions_rms[rms]()
+    bkg = Background2D(
+            data,
+            boxs,
+            filter_size=filters,
+            sigma_clip=SigmaClip(sigma=sigma),
+            bkg_estimator=bkg_estimator,
+            exclude_percentile=exclude_percentile,
+            bkgrms_estimator=bkgrms_estimator,
+            mask=mask_source)
+    verboseprint("Mask, median = %0.2f" % (bkg.background_median))
+    verboseprint("Mask, rms = %0.2f" % (bkg.background_rms_median))
+    # if i == 0:
+    fitsfile.data = fitsfile.data - bkg.background  # .astype('uint16')
+    # else:
+    #     fitsfile.data[np.isfinite(mask)] = fitsfile.data[np.isfinite(mask)] - bkg[-1].background[np.isfinite(mask)]  # .astype('uint16')
+    # if len(masks) == 2:
+    #     masks[-1][np.isfinite(masks[-1])] = fitsfile.data[np.isfinite(masks[-1])]
+    # if len(masks) == 2:
+    #     diff = np.nanmean(fitsfile.data[np.isfinite(masks[1])]) - np.nanmean(fitsfile.data[np.isfinite(masks[0])])
+    #     fitsfile.data[np.isfinite(masks[1])] -= diff
+    #     diff = np.nanmean(np.hstack((fitsfile.data[np.isfinite(masks[0])], fitsfile.data[np.isfinite(masks[1])])))
+    #     fitsfile.data[np.isfinite(masks[1])] -= diff
+    # else:
+    # diff = np.nanmean(fitsfile.data[np.isfinite(masks[0])])
+    # fitsfile.data[np.isfinite(masks[0])] -= diff
+    # if np.isfinite(data).all():
+    #     fitsfile.data = fitsfile.data.astype("uint16")
 
     name = os.path.join(os.path.dirname(filename) + "/bkgd_photutils_substracted/%s" % (os.path.basename(filename)))
     fitswrite(fitsfile, name)
@@ -9427,6 +9435,7 @@ def python_command(xpapoint=None,argv=[]):
 def Button(xpapoint=None):
     """ Creates a pyQt5 button to continue the different tutorial on DS9 or quit
     """
+    #Needs to be installed via mamba install PyQt to be sure it works on OSX, and on linux: sudo apt-get install libxcb-xinerama0
     from PyQt5.QtWidgets import QApplication, QWidget, QVBoxLayout, QPushButton, QLabel
     from PyQt5.QtGui import QIcon, QPixmap
     from PyQt5.QtCore import pyqtSlot  # <--- add this line

diff --git a/pyds9plugin/QuickLookPlugIn.ds9.ans b/pyds9plugin/QuickLookPlugIn.ds9.ans
@@ -466,7 +466,7 @@ hmenu 3. Astronomical Softwares
 		Subtract Background
 		*
 		menu
-		DS9Utils background_estimation -x $xpa_method  $param(Background-Subtraction); -s $sigma -b $bckd -r $rms -f $filter -box $box -per $percentile -m $mask -snr $snr -n $npixels -d $dilate -p -  | $text
+		DS9Utils background_estimation -x $xpa_method  $param(Background-Subtraction); -s $sigma -b $bckd -r $rms -f $filter -box $box -per $percentile -m $mask -snr $snr -n $npixels -d $dilate   | $text
 
 		Aperture photometry
 		*

diff --git a/pyds9plugin/QuickLookPlugIn_DS9<8.2.ds9.ans b/pyds9plugin/QuickLookPlugIn_DS9<8.2.ds9.ans
@@ -421,7 +421,7 @@ hmenu 3. Astronomical Softwares
 		Subtract Background
 		*
 		menu
-		DS9Utils background_estimation -x $xpa_method  $param(Background-Subtraction); -s $sigma -b $bckd -r $rms -f $filter -box $box -per $percentile -m $mask -snr $snr -n $npixels -d $dilate -p -  | $text
+		DS9Utils background_estimation -x $xpa_method  $param(Background-Subtraction); -s $sigma -b $bckd -r $rms -f $filter -box $box -per $percentile -m $mask -snr $snr -n $npixels -d $dilate  | $text
 
 		Aperture photometry
 		*

diff --git a/setup.py b/setup.py
@@ -35,7 +35,7 @@
 
 MAJOR = '3'
 MINOR = '0'
-MICRO = '1dev5'
+MICRO = '1dev6'
 version = '%s.%s%s' % (MAJOR, MINOR, MICRO)
 
 
@@ -69,7 +69,7 @@ def setup_package():
     version=version,
     license='CeCILL-B',
     install_requires = requires,
-    long_description= " A python DS9 extension for quicklook processing of astronomical images. This highly interactive extension can be generalized automatically to a set of images to turn the plug-in into a real multi-processing pipeline.",
+    long_description= long_description,#" A python DS9 extension for quicklook processing of astronomical images. This highly interactive extension can be generalized automatically to a set of images to turn the plug-in into a real multi-processing pipeline.",
     url='https://people.lam.fr/picouet.vincent/index.html',
     platforms=["Linux", "Mac OS-X", "Unix"],
     cmdclass={'install': PostInstallCommand,'develop': PostDevelopCommand},