I am glad to announce that PixInsight 1.8.7 has been released today. This version is a particularly important milestone, and I'm proud of all the hard work we have put into it.
Installation packages for PixInsight 1.8.7 are now available to all licensed users for FreeBSD, Linux, macOS and Windows:
https://pixinsight.com/dist/This version is a major update that you have to install manually. It is not available as an application update.
If you have version 1.8.6 already installed, this new version does not require a license reactivation. If you are still using a 1.8.5 version or older, see the
official 1.8.6 version announcement for a detailed description of our new licensing system. See also
FAQ 2.10 for additional information on license reactivations.
New FeaturesAs always, our releases come with new features, bug fixes, and improvements. Here is a brief list with the most important changes.
PixInsight Platform: New Versioning SystemSo far PixInsight versions have consisted of four elements: major, minor, release and build version numbers. For example: 1.8.6.1475. Since version 1.8.7 onward, build version numbers will no longer be used. This applies to all components of the PixInsight platform, including the core application, all officially released modules, and our C++ and JavaScript development frameworks. So from now on you'll see new PixInsight versions released as 1.8.7, ..., 1.8.42, ... and so on.
For compatibility with some existing scripts and modules, the core application reports a
fixed build number when requested from C++ or JavaScript code, which is higher than all previous versions. However, new scripts
should not rely on build numbers anymore to check for existence of core features or other purposes.
Core Application: New Workspace SelectorsOne of the most visible changes in the PixInsight core 1.8.7 application is its new Workspace Selector tool bar. Workspace selector items are now dynamic components that provide visual information on workspace contents. You'll see three small dots drawn on a selector with the following meaning:
- Top dot: The workspace contains process interface windows.
- Middle dot: The workspace contains process icons.
- Bottom dot: The workspace contains images.
The new workspace selectors provide you real-time feedback as your data and processes are being distributed throughout workspaces. This feature is particularly useful when you open a project. No more guessing about where icons, tools and images are located; now you know that at a glance.
Core Application: Improved Transfer View Zoom/Position FeatureWhen you drag a view selector and drop it on another selector, the position and zoom level of the source view are transferred to the target view. So far this only worked when both views had the same dimensions. Since version 1.8.7 this always works, irrespective of view geometry differences. When the views have different dimensions, target positions and zoom ratios are computed to approximate the source view conditions as best as possible. This is very useful to compare images at different scales, such as drizzle x2 and normal integration results.
StarAlignment: New Arbitrary Distortion Correction AlgorithmA considerable amount of research and development work has been done on the StarAlignment tool in PixInsight 1.8.7. The most important changes are the following:
*
A new arbitrary distortion correction algorithm, redesigned and reimplemented from scratch. This two-stage algorithm (iterative global distortion and local distortion models) aims at achieving centipixel accuracy in most image registration tasks.
*
Star positions and fluxes are now computed from robust PSF fittings, including a new statistical weighting system based on robust goodness-of-fit estimates. This largely improves reliability and accuracy of image registration, especially under strong distortions.
*
Improved intersection computation, which is now more efficient, accurate and robust. Frame intersections are computed by default in mosaic generation modes, and can also be enabled manually in register/match mode to solve small overlapping issues automatically.
*
Improved RANSAC model validation routines, which are now more efficient and resilient to ill-posed image registration problems.
*
Generation of distortion maps, which are special control images where each pixel value is proportional to the difference between the applied image registration model and a projective transformation. These maps are useful to check and evaluate image distortions.
*
New randomized mosaic combination feature. In register/union mosaic mode, instead of replacing reference image pixels with registered target image pixels, both images are now combined by selecting pixels from one of the images randomly. This works much better to produce seamless mosaics without visible texture differences.
*
Improved star maps generated in detected/matched stars working modes. Star maps can now be used directly as masks.
*
Metadata items describing center celestial coordinates and image scale are now replaced with reference image values in all registered target images. This includes RA/DEC, OBJCTRA/OBJCTDEC, XPIXSZ/YPIXSZ and FOCALLEN FITS keywords, as well as the corresponding Observation:Center:RA, Observation:Center:Dec, Instrument:Sensor:XPixelSize, Instrument:Sensor:YPixelSize and Instrument:Telescope:FocalLength XISF properties.
*
The user interface has been simplified. Many advanced parameters have been removed and are only available by editing instance source code. This has been possible because the new distortion correction algorithm is much more robust and requires virtually no manual intervention. It uses much more parameters internally, but there is no reason to change their default values, except in very unusual cases.
ImageIntegration: New Winsorization Cutoff Parameter / Improved Linear Fit ClippingSeveral improvements have also been implemented in the ImageIntegration tool. Besides other minor performance improvements, these are the main changes:
*
New automatic buffer sizes feature. When this option is enabled, the ImageIntegration tool uses the largest possible
buffer size and
stack size values, calculated automatically from the amount of free physical memory currently available to PixInsight. Usually this is the best option to optimize image integration performance, especially because it is completely automatic. This option is enabled by default.
*
New truncate on out-of-range parameter. When the output integrated image has saturated pixel samples out of the nominal [0,1] range, this option truncates saturated pixels instead of rescaling the whole image. No out-of-range values should occur after integration of a well-calibrated data set. However, with improperly calibrated data, saturated pixels may lead to out-of-range values after output normalization, depending on the frame selected as integration reference. When this happens, the best option for integration of light or science frames is a linear rescaling, which preserves all of the integrated data. However, in some cases altering all pixel values is not admissible, so a rescaling operation is not applicable. This is the case for integration of flat frames, where truncation is the only option available to preserve the correct illumination profile in the integrated master flat frame.
*
New subtract pedestals option. When this option is enabled, existing (nonstandard) PEDESTAL keyword values are subtracted from input images. Some applications add small positive values (typically about 100 DN) systematically to calibrated light frames. These small
pedestals should be subtracted from source integration pixels to refer all input data to the same zero point consistently. This option should be enabled under normal working conditions, and hence is enabled by default. PEDESTAL keywords generated by PixInsight always have positive values. Other applications may write negative values (another 'nice' example of FITS interoperability issue). For improved compatibility, when ImageIntegration finds a negative pedestal, it issues a warning message and assumes that the value is intended to be added.
*
New Winsorization cutoff parameter. This is a cutoff point for the Winsorized sigma clipping rejection algorithm, expressed in sigma units. All pixel samples with absolute differences from the median larger than this parameter will be set equal to the median of the pixel stack in the first rejection iteration. This replaces extreme outliers, such as cosmics, hot and cold pixels, with plausible values instead of their nearest neighbors. This is more correct because allowing some influence from severe outliers does not make sense---after all, we know they are outliers without uncertainty, so Winsorizing them is a conceptual error. These errors, present in previous versions of the ImageIntegration tool, are insignificant statistically, so their practical influence is negligible. Nevertheless, we now have a fully correct implementation of Winsorized sigma clipping since PixInsight 1.8.7. The default Winsorization cutoff parameter value is 5 sigmas, which should be sufficient to detect extreme outliers without compromising the performance of the algorithm in terms of preservation of significant data.
*
Improved implementation of linear fit clipping rejection. So far we have been using a slight simplification of the linear fit clipping algorithm. Instead of using the distance from each pixel value to the fitted straight line to make rejection decisions, we used only the vertical distance, for performance reasons. We now use accurate distance values, which is more correct mathematically, with a small computational cost. Again, the errors made in previous versions were negligible, mainly because fitted line slopes are usually small (for a nearly horizontal line, the vertical distance is a good approximation to the true distance from a point to the line). As a result of this change, you may notice small changes in the way linear fit clipping works, in terms of the required low and high clipping points, hopefully leading to a better rejection with more preservation of significant data.
*
Improved preservation of existing metadata. Center celestial coordinates and geodetic observation coordinates are now averaged from the set of input images, and a large set of XISF properties and FITS keywords are preserved in the output integrated image. This greatly facilitates image analysis tasks, mainly for astrometry and photometry.
DrizzleIntegration: Improved Integration Geometry and Pixel Rejection / Local Distortion Support*
Improved integration geometry. Previous versions of the DrizzleIntegration tool had minor issues with the geometry of the final integrated image. As a result, drizzle-integrated and normal integrated images were not perfectly registrable, with differences up to 0.5 pixels. All of these issues have been fixed in the new version. The results of DrizzleIntegration and ImageIntegration for the same data set are now directly registered to centipixel accuracy, with an integer scaling factor equal to the applied drizzle scale.
*
Improved pixel rejection. Previous versions of DrizzleIntegration made small pixel rejection mistakes sometimes. This has been fixed.
*
DrizzleIntegration fully supports local distortion models generated by StarAlignment, which are now stored in .xdrz files. A new
enable local distortion parameter has been added to provide control over this feature.
*
Improved preservation of existing metadata, with the same functionality implemented in ImageIntegration.
PhotometricColorCalibration: New Narrowband Working ModeThe PhotometricColorCalibration tool implements a new
narrowband working mode, along with the broadband mode already available in previous versions. For clarification, let's describe both modes and its associated working parameters:
*
Broadband working mode: PCC calibrates the color of a picture acquired with broadband RGB filters. In this mode, the color calibration process equalizes the emissions of the specified white reference in the RGB filters. This means that, after performing the color calibration, the entire light coming from the white reference would have a proportion of 1:1:1. We base the color calibration on the photometry of the stars detected in the image. This photometry indicates the RGB weights needed to neutralize a white reference model, so we actually don't need an instance of the white reference inside the image.
*
Narrowband working mode: Calibrate the color of a picture acquired with narrowband filters. Each color channel of the image should contain a single-waveband filter. This function equalizes the photons of the nebular emission lines in the different filters. As a result, the proportion of the emission line strengths, as emitted by the object, is preserved in the image. The wavelength and bandwidth of each narrowband filter should be specified, and to this purpose PCC has a new set of six parameters to define these values for the filters used in the red, green and blue channels of the target image. The default parameter values correspond to the frequently-used HOO combination (H-alpha for red and O-III for green and blue).
Both working modes implement algorithms designed by PTeam member Vicent Peris.
ImageSolver Script: New Advanced Alignment ModePixInsight 1.8.7 includes the latest version 5.4.1 of the ImageSolver script, written by PTeam member Andrés del Pozo. This version implements a new algorithm to model field distortions more accurately, especially on wide field images. Basically, when this special mode is enabled the solver evaluates field distortions separately on nine image regions distributed on a regular 3x3 grid. This generates more accurate distortion models, especially for strongly distorted images acquired with consumer short focal length lenses. Along with this improvement, ImageSolver includes small changes necessary to adapt the script to the new features implemented in the StarAlignment tool.
PixelMath: New rndselect FunctionThe rndselect function is formally described as:
rndselect( a, b[, ...] )
This function evaluates to a randomly selected value among the set of arguments {a, b, ...}. This function is fantastic to combine images without visible texture differences. For example, to combine two separate mosaic frames A and B, you can use the following expression:
iif( A && B, rndselect( A, B ), A + B )
This function has been borrowed from the new MosaicGenerator tool, where I am implementing a randomized multiscale sub-band blending algorithm.
Updated Databases and Third-Party Support LibrariesPixInsight 1.8.7 includes the following updates to critical data sources and support libraries:
* The RAW format support module integrates the latest
LibRaw-201903 snapshot version.
* Integration of
gnuplot version 5.2.7* Core application built with the latest
Qt version 5.12.5.
*
Delta T database updated with the latest data available from IERS Rapid Service/Prediction Center online files, as of 2019 September 29. Delta T is the difference in seconds between the TT (Terrestrial Time) and UT1 timescales for a given date. It is an essential observational quantity for ephemeris calculations and reduction of astronomical positions.
*
CIP-ITRS database updated with the latest data available from IERS Rapid Service/Prediction Center online files, as of 2019 September 29. This database provides the coordinates of the Celestial Intermediate Pole (CIP) referred to the International Terrestrial Reference System (ITRS). These coordinates are necessary to account for polar motion in calculation of topocentric coordinates of solar system objects.
Bug FixesPixInsight 1.8.7 fixes all confirmed bugs detected since the last 1.8.6 release, with a unique exception: a
complex bug involving the PixelMath and Statistics tools, which still is present in this version. I'll try to fix this problem as soon as possible with an update to the core application.
Known ProblemsFreeBSD* Recent versions of the KDE Plasma desktop environment include a
morphing popups desktop effect that causes problems with PixInsight 1.8.6 and 1.8.7. If this effect is enabled, it slows down some dragging operations in the PixInsight core application, such as dragging icons and view selectors. You should disable this effect, which is very easy from KDE's System Settings panel. This problem is also present on Linux.
Linux* PixInsight is not compatible with the open-source Nouveau graphics driver. If your distribution is using Nouveau, you must remove and replace it with the proprietary Nvidia driver in order to use PixInsight.
* The same problem with KDE's
morphing popups desktop effect exists on FreeBSD and Linux (see above).
macOS* On Apple laptops with dual graphics cards, the integrated Intel graphics driver causes problems with PixInsight 1.8.6 and 1.8.7. If you use one of these laptops, we recommend you disable the
Automatic graphics switching option on System Preferences (Energy Saver section) before running the PixInsight core application. Otherwise you may experience some interface usability problems.
Windows* There are problems with Intel graphics drivers on machines where the only graphics card available is an integrated Intel HD graphics GPU. This mainly affects laptops. This does not mean that PixInsight cannot work on these machines, just that you can expect screen rendering and usability issues. Unfortunately, there is nothing I can do to solve these problems, for now.
* On some laptops with nonstandard screen resolutions, such as the Microsoft Surface, PixInsight cannot find valid screen scaling factors automatically. On these machines, you may have to run the PixInsight core application with the --ui-scaling command line argument. For example, to apply a UI scaling factor of 2:
C:\Program Files\PixInsight\bin\PixInsight.exe --ui-scaling=2
You only have to do this once, since the scaling factor will be stored in application preferences automatically. Of course, if you already are using version 1.8.6 and the interface looks correct, then you don't need to do this; this is only required for new installations.
* The Export as PDF feature of the integrated web browser component does not work on Windows. I'll try to fix this in a future release.
Recommended Platforms and Supported Operating SystemsThe reference implementations of PixInsight 1.8.7, where you may expect the maximum performance and best user experience, are the FreeBSD and Linux versions. Currently our primary development platforms are:
-
Kubuntu Linux 18.04 LTS-
FreeBSD 12.0On FreeBSD and Linux we strongly recommend running PixInsight on the KDE Plasma desktop environment. PixInsight 1.8.7 has been tested on the following Linux distributions:
Kubuntu 18.04 LTS
Kubuntu 19.04
Fedora 29
Fedora 30
Linux Mint 19.1
PixInsight 1.8.7 for Linux requires GLIBC 2.27 or a newer version.
PixInsight 1.8.7 for macOS is only compatible with macOS 10.12, 10.13 and 10.14. Older versions are not supported. The next macOS 10.15 should be supported without problems, although this has not been tested yet. Our installation packages have been notarized by Apple.
PixInsight 1.8.7 for Windows is compatible with Windows 10 exclusively. Windows 8.1, Windows 7 and older versions are not supported. The application
may run well on Windows 7, but there is no guarantee.
Hardware Requirements: Breaking ChangesThe FreeBSD, Linux and macOS versions of PixInsight 1.8.7 require a processor with SSE4.2 instruction support.
Short-Term Development LinesDuring the next months we plan on releasing the following new products on the current PixInsight 1.8 platform:
* MosaicGenerator. An astrometry based automatic mosaic construction tool.
* DynamicBackground. An improved background modeling/correction tool, which will be the successor to the current DBE tool.
* A new tool (name still unknown) for astrometry based measurements, integrated with the core application.
* A new high-performance format for storage of large star databases. The Gaia DR2 catalog will be available for download as one of these databases. This will greatly improve all of our astrometry based tools, such as the ImageSolver script and the new MosaicGenerator tool, which will be able to work with local files without the limitations imposed by online services.
___________________
Thank you for your attention and continued support. Enjoy!
