Astro Pixel Process...
 
Share:
Notifications
Clear all

15th Feb 2024: Astro Pixel Processor 2.0.0-beta29 released - macOS native File Chooser, macOS CMD-Q fixed, read-only Fits on network fixed and other bug fixes

7th December 2023:  added payment option Alipay to purchase Astro Pixel Processor from China, Hong Kong, Macau, Taiwan, Korea, Japan and other countries where Alipay is used.

 

[Closed] Astro Pixel Processor 1.075

1 Posts
1 Users
0 Likes
5,763 Views
(@mabula-admin)
Universe Admin
Joined: 7 years ago
Posts: 4366
Topic starter  

Astro Pixel Processor 1.075

  • IMPROVED, 4) REGISTER, PROJECTIONS OF DATA WITH FIELD OF VIEWS GREATER THAN 120 DEGREES, a big improvement has been made for registration of very wide field of views. You can now choose Rectilinear, Equireactangular and Mercator projections if you used the calibrated projective registration model. In addition, these projection methods allow you to change the Center of Projection (COP) of your data. You can move the COP horizontally and vertically. If you have very a very wide Field Of View (FOV), it's best to not use the rectilinear projection, because it only works for FOVs smaller than 120 degrees. Here is the same data projected differently:

180 degree milkyway mosaic (courtesy of Ralph Wagter):

top row: Rectilinear, Equirectangular, Mercator projections

middle row: the same but with the COP moved 30 degrees to the top of the FOV

bottom: Equirectangular, Mercator projections with the COP moved 30 degrees to top and right side of the FOV

APP projections
Projection settings of data
  • IMPROVED, 4) REGISTER, DIRECTLY ROTATE FIELD OF VIEW, you can now directly rotate the field of view of your integration/stack by setting the projection rotate slider. After registration, see your field of view of the registered data with the l-c-registered image viewer mode. Then adjust the rotate slider and view again, you will see that the field of view rotates. This is possible with both projective and calibrated projective models.
  • FIXED, ADOBE DNG FROM PENTAX CAMERA SUPPORT, there was a metadata issue, which prevented APP from properly loading Pentax camera produced ADOBE DNG frames. This is fixed.
  • ADDED, Canon CR2 support, added support for the EOS 4000D/REBEL T100, which are the same as the EOS 3000D
  • UPDATED, LIBRAW, LibRaw for read support of Sony ARW, Adobe DNG and soon to come other camera brands has been updated to 0.19.5, the current latest stable release. Bear in mind that APP uses it's own raw conversion code for Canon CR2 and Nikon NEF frames.
  • ADDED, ADOBE DNG SUPPORT, APP now supports Adobe DNG frames, so all frames of yet unsupported camera brands can be converted with Adobe DNG Converter and then the resulting dng's can be processed with APP. DNG conversion has been tested for raw data of camera models of the following brands, Canon, Nikon, Sony,Kodak, Hasselblad,Panasonic, Pentax, Leica, Olympus. Please use the latest Adobe DNG Converter available here: https://helpx.adobe.com/photoshop/digital-negative.html#downloads
  • FIXED/IMPROVED, 3) STAR ANALYSIS, the star analysis module now fully supports severely undersampled data. Undersampled data means that the star sizes in Full Width Half Maximum (FWHM) are less than 2 pixels. APP will now be able to process these frames as well. Undersampling could occur due to having too large pixels for your focal length and atmospheric conditions, or due to binning of your camera's sensor for instance.
  • FIXED, BAYER CFA RECOGNITION SHARPCAP capture software, APP should now automatically detect that the data is Bayer CFA data if captured with SharpCap and should know the correct Bayer CFA pattern through reading the BAYERPAT Fits header tag that SharpCap uses. SharpCap reports the CFA pattern based on reading the data from bottom to top, left to right, whereas APP reads the data from top to bottom, left to right. APP will now automatically correct the reported Bayer CFA pattern in the Fits metadata. So a pattern of "GBRG" is automatically interpreted as "RGGB". See screenshot, SharpCap does notify that the pattern could be different because of the way software would read the fits data.
SharpCap BayerCFA pattern
  • FIXED/IMPROVED, COSMETIC CORRECTION in 2) CALIBRATE, Hot Pixel correction in 1.075.beta4 had a severe Memory Leak and it was slow as well. The memory leak is gone and the hot pixel correction is much faster now. In addition, the bad column correction and detection have both been improved further giving more robust correction.
  • FIXED/UPGRADED, NEW OPENGL 2/3/4 IMAGE VIEWER, the new OpenGL Image Viewer should now work on almost all systems depending on graphical drivers that support OpenGL.
  • FIXED, CALIBRATION OF BAYER CFA DATA WITH UNEVEN DIMENSIONS, as reported by Henry @minusman, there was a bug in creation of the Masterflat when your data has uneven dimensions and is Bayer CFA. This is now fixed:
Fixed BayerCFA calibration uneven dimensions 1
  • FIXED, BAYER CFA RECOGNITION ZWO ASIAIR, APP should now automatically detect that the data is Bayer CFA data if captured with the ZWO ASIAIR and know the correct Bayer CFA pattern through reading the BAYERPAT Fits header tag that the ZWO ASIAIR uses.
  • IMPROVED/NEW FEATURE, COSMETIC CORRECTION OF HOT & COLD COLUMNS & HOT PIXELS, in this new version, the hot/cold column detection in the Bad Pixel Map has been made redundant by now providing correction of these bad columns directly on the (calibrated-)light frames. This proved to work much better in the end and the algorithm used now  is very robust and simple to use. Bad columns are mostly a problem on CCD camera's and we have tested it extensively on several CCD camera's of APP users and the CCD camera's of iTelescope.net. In addition you can now also correct hot pixels cosmetically besides using the Bad Pixel Map functionality off course. Bad Pixel Maps created with older APP versions will still work the same, but if you are using them to correct bad columns, we strongly recommend that you create new Bad Pixel Maps only for the hot/cold pixels and correct the bad columns that you have with the new cosmetic correction feature, which works really very easy and robust. In the screenshots below you will see a nice example of the new Cosmetic Correction functionality on Hydrogen Alpha data of telescope T33 of iTelescope.net.

Uncalibrated Hydrogen-Alpha capture of NGC2070, the Tarantula Nebula shot with iTelescope's T33, a 0.3m reflector telescope with an Apogee Alta U16 CCD camera, please check the hot pixels and the bad columns present on this uncalibrated light frame:

Cosmetic Correction uncalibrated frame
Cosmetic Correction raw T33 astropixelprocessor NGC 2070 20190910 011756 Ha BIN1 E 300 003 St

The same Hydrogen-Alpha capture,but now calibrated with a MasterBias, MasterDark and MasterFlat, no Bad Pixel Map was used, but the new Cosmetic Correction was enabled to correct remaining hot pixels and both hot and cold columns.

Cosmetic Correction calibrated frame
Cosmetic Correction calibrated T33 astropixelprocessor NGC 2070 20190910 011756 Ha BIN1 E 300 003 St

Both the uncalibrated and calibrated image in one view. Please note that the corners are not well corrected in this case, due to not properly created flats which is not a problem here at all, to illustrate the new bad column and hot pixel Cosmetic Correction functionality:

T33 uncalibrated versus calibrated hydrogen alpha capture

The last screenshot shows the functionality in 2) Calibrate, please note, that the bad column functionality for the Bad Pixel Map has been removed completely in favor of the Cosmetic Correction:

Cosmetic Correction new Functionality
  • IMPROVED/UPGRADED, OPENGL IMAGE VIEWER, added GL 3 and GL4 support, the previous OpenGL implementation was only based on the OpenGL 2 interface. This interface is old and uses a so-called Fixed Function Pipeline. Some GPUs and drivers no longer support this specification by both hardware and software so on those computers, APP could not use OpenGL. APP has now been upgraded with both OpenGL3 and OpenGL4 support. These interfaces use the so-called Programmable Shader Pipeline, which is more low-level and better suited to be used with newer GPUs. So this upgrade should give better image viewer performance on a wide range computers depending on the installed graphical hardware and it's software drivers.
OpenGL upgrade GL3 GL4
  • FIXED, OSC/COLOR/BAYER CFA DATA CALIBRATION, multipe bugs were solved which manifested when you tried to calibrate data from a camera with a Bayer CFA sensor which has uneven dimensions. The issue was reported by a user of a CANON REBEL T7/1500D/KISS X90 camera. This camera has a sensor with uneven height. The bug fixes were also applied to all other image loaders (Sony, Nikon, Fits, Tiff, jpeg) and all modules that are related to Bayer CFA processing.
  • FIXED, CALIBRATION, making an artificial masterflat using the smooth model (last masterflat blur option) was not working. Now works. Please realise, that such a smooth masterflat will be based on a Kang-Weiss model. It does not correct dust, nor will it correct odd illumination patterns from for instance a OAG or a DLSR sensor chamber.
  • IMPROVED, APPLICATION SECURITY, several adjustments were made relating to application security, license activation and verification with our license server. Sometimes, when APP starts and communication with the license server is slow for whatever reason, application start-up time could suffer. The application is now first fully started, and license activation/verification is done when the user interface is already visible which should improve application start-up experience. Furthermore, deactivating an activated computer, using the top left LICENSE button, will no longer freeze the application until the de-activation is finished. All communication with our license server is now done on background threads.
  • FIXED, SORTING/ANALYTICAL GRAPH OF FRAMES AFTER STAR ANALYSIS, in APP 1.074(.1) the sorting function and analytical graph of the frames for star shape and quality did not work directly after star analysis. After registration, it did work. This is now fixed.
FIX StarAnalysis Sorting AnalyticalGraph

 

  • FIXED, INTEGRATION OF SINGLE FRAME DATA, if you would load 1 frame into APP then Integration would not complete, saying you need at least 2 frames to be able to make an integration. Now with the Multi-Channel/Session features, this no longer makes sense. So if you have only 1 frame per channel/session, APP will now still process and complete this task.
  • FIXED, NORMALIZATION using an unselected REFERENCE frame, APP would give an Error message: java.lang.Float cannot be cast to java.lang.Double if you start normalization with an unselected reference frame. This is now fixed. 
  • FIXED, RGB Combine luminance channel recognition, if you load a luminance channel into the RGB Combine tool it will now automatically recognize and suggest to assign it to Luminance.
  • FIXED, MacOS FREE HDD SPACE, as reported in this topic https://www.astropixelprocessor.com/community/macos/incorrect-available-storage-in-macos-potential-bug/ APP has a hard time to know the exact free HDD space available for an integration task due to the so-called purgeable free space introduced by Apple since MacOS Sierra. Previously, the integration task would simply be cancelled although the actual free HDD space available would be sufficient to complete the task...

    When APP thinks that there is not enough space for an integration task, APP will no longer simply cancel the task, but it will give the user 4 choices 😉

    1. Before continuing with the integration, you can free HDD space yourself and then continue.
    2. Let APP test the actual HDD space requirement by creating a temporary file of 105% of the space needed for the integration. This goes very fast and it will actually test if the integration would be possible... this is the solution for APP users that are plagued by the purgeable free space on MacOS.
    3. just ignore the warning and let APP continue...
    4. cancel the integration task

(please disregard the actual numbers for freespace in the image below, it's a popup triggered by testing this)

MacOS purgeable Free Space Solution
  • IMPROVED, CALIBRATION ENGINE, SEPARATE DARKS ACCORDING TO EXPOSURE TIME, a new setting has been added to 2) CALIBRATE, the calibration engine. You can now elect to bundle all of your darks of different exposure times into 1 masterdark or have masterdarks created according to the different exposure times. Realise that darks of different ISO/gain, filters and/or sessions, are always treated separately to create different masterdarks. The setting is called: separate darks acc. to exposure time (see screenshot) and will be enabled by default.
Separate Darks According To Exposure Time
  • FIXED, MULTI-CHANNEL/FILTER & -SESSION, in some cases, a multi-channel/filter and/or multi-session integration would not show the proper frame marks in the frame list panel frame column. This should now be correct all the time.
  • IMPROVED, CALIBRATION, MASTERDARK/MASTERDARKFLAT MATCHING WITH LIGHTS/FLATS, APP will now always automatically match MasterDarks/MasterDarkFlats with Light/Flat frames if a MasterDark/MasterDarkFlat is present within 75% of the needed exposure time for correct calibration. If there are more Masters present within this margin, the closest one with respect to exposure time is used. If there are no Masters within this margin, APP will ask you which one to use only if there is more than 1 Master present. If there is only 1 present, it is automatically chosen anyway.
  • IMPROVED, STAR ANALYSIS, further improvements have been implemented in star detection to make it more robust on both data with small and large FWHM values. Especially discrimination between hot pixels and small stars has been further improved. Further improvements were triggered by this topic where registration was done on hot pixels instead of stars: https://www.astropixelprocessor.com/community/main-forum/my-stars-wont-align-after-integration/
  • IMPROVED, SORTING OF FRAMES IN FRAMELIST PANEL, there was a small bug that would switch the order of the frame types when you flipped the sorting on analytical data like the star shape. So then the Calibration Masters would be at the top of the panel and the light frames as the bottom... that is fixed now. In addition, the sorting order is no longer called flip or unflip, but has been called: sort frames best to worst or sort frames worst to best
  • IMPROVED, NOISE REDUCTION INFORMATION IN INTEGRATION FITS HEADER, several comments are added into the FITS headers to clarify noise reduction metrics of your light frame integrations. In addition, a couple of new header tags are added as well, giving even more information about the noise reduction that you get with integration of your light frames:
Noise Reduction Metrics Fits Header
  • IMPROVED, IMAGE LOADER, the first time that you load images after APP has started, APP will simply ask you if you need to enable the Multi-Channel/Filter and/or Multi-Session processing modes to be able to process your data properly.
  • IMPROVED, STAR ANALYSIS, STAR SHAPE METRIC, in the frame list panel you will now see the star shape metric values reported as well, besides the Full Width Half Maximum (FWHM) values. The star shape metric is a combination of the star size & star roundness in your images. The higher the score, the better. Frames with smaller and more circulair stars will have higher scores than frames with bigger and less circulair stars :
StarShape metric added to analytical results
  • IMPROVED, FRAME LIST PANEL, MULTI-FRAME SELECTION WITH SHIFT & CONTROL KEYS, you can now do multi-frame selection/deselection/remove/delete using both the shift and control keys combined with the right-mouse click popup menu:
Multiple Frame Selection With CTRL SHIFT keys
  • IMPROVED, 2) CALIBRATE, DARK SCALING SELECTION, the selectboxes to enable/disable dark frame scaling on your MasterDarkFlats or MasterDarks are moved downwards in the 2) CALIBRATE menu, because dark frame scaling should be considered as a general calibration setting. They don't affect how the MasterDarkFlats or MasterDarks are created:
Dark Scaling general calibration setting
  • FIXED, IMAGE NORMALIZATION, CALIBRATED MAXIMDL FRAMES, APP is now able to properly show and process frames that are 32bits floats from MaximDL. MaximDL creates 32bits frames without metadata about how the images could/should be scaled in 32bits floats. Normally, the 32bits float data is put in the 0-1 range, or left in the 16bit range of 0-65535, but MaximDL does not adhere to any range it seems.
  • FIXED, MASTERS, DOUBLE INPRECISION GAIN/EGAIN values, some applications like MaximDL report gain/egain values with machine inprecision errors:

HDU1 - EGAIN = 1.7200000286102295 / Electronic gain in e-/ADU
HDU1 - SWCREATE= 'MaxIm DL Version 5.24 140126 0KJR5' / Name of software that created

APP will now correct for this inproper reporting in the naming of the masters and APP's metadata:

HDU1 - SOFTWARE= 'Astro Pixel Processor by Aries Productions' / software
HDU1 - VERSION = '1.075-beta2' / Astro Pixel Processor version
HDU1 - CALFRAME= 'MasterDark' / master dark frame
HDU1 - INSTRUME= 'SBIG STL-11000 3 CCD Camera' / instrument name
HDU1 - CFAIMAGE= 'no ' / Color Filter Array pattern
HDU1 - GAIN = 1.72 / gain or ISO depending on instrument

And the master will be:

MD-ISO_gain_1.72-exp_60.0s-10subs-SBIG_STL-11000_3_CCD_Camera-2004x1336-SC_1_3.0-avg.fits

instead of:

MD-ISO_gain_1.7200000286102295 -exp_60.0s-10subs-SBIG_STL-11000_3_CCD_Camera-2004x1336-SC_1_3.0-avg.fits

  • FIXED, USING MULTIPLE MONITORS WITH DIFFERENT SURFACE SCALES, when you attached regular external monitors to, for instance, a MacBook Pro with a Retina display, the image drawing with OpenGL on the external monitors would not be correct. The issue was caused by APP not recognising that APP would be used on another screen with a different Surface Scale. Retina displays use a Surface Scale of 2.0, whereas regular monitors only use 1.0. Now, when you drag APP across your desktop workspace to another screen, APP will automatically adjust and thus will show your images correctly on all external monitors attached. The screen shot shows 3 Astro Pixel Processor instances on 3 different screens. To the right is a Retina display of a MacBook Pro, the 2 other instances are shown on 2 different external monitors, of different brands, attached to the MacBook:
Screenshot 2019 09 13 at 20.41.02

 

This topic was modified 5 years ago 52 times by Mabula-Admin
This topic was modified 4 years ago by Mabula-Admin

   
Share: